Affymetrix 10-K 2007
Documents found in this filing:
x ANNUAL REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934
FOR THE FISCAL YEAR ENDED DECEMBER 31, 2006
o TRANSITION REPORT PURSUANT TO SECTION 13 OR 15(d) OF THE SECURITIES EXCHANGE ACT OF 1934 (NO FEE REQUIRED)
FOR THE TRANSITION PERIOD FROM TO
COMMISSION FILE NUMBER 0-28218
(Exact name of registrant as specified in its charter)
(Registrants telephone number, including area code)
Securities registered pursuant to Section 12(b) of the Act:
Securities registered pursuant to Section 12(g) of the Act:
Indicate by check mark if the registrant is a well-known seasoned issuer, as defined in Rule 405 of the Securities Act. Yes o No x
Indicate by check mark if the registrant is not required to file reports pursuant to Section 13 or Section 15(d) of the Act. Yes o No x
Indicate by check mark whether the registrant (1) has filed all reports required to be filed by Section 13 or 15(d) of the Securities Exchange Act of 1934 during the preceding 12 months (or for such shorter period that the registrant was required to file such reports), and (2) has been subject to such filing requirements for the past 90 days. Yes x No o
Indicate by check mark if disclosure of delinquent filers pursuant to Item 405 of Regulation S-K is not contained herein, and will not be contained, to the best of registrants knowledge, in definitive proxy or information statements incorporated by reference in Part III of this Form 10-K or any amendment to this Form 10-K. x
Indicate by check mark if the registrant is a large accelerated filer, an accelerated filer, or a non-accelerated filer. See definition of accelerated filer and large accelerated filer in Rule 12b(2) of the Exchange Act. (Check one).
Indicate by check mark whether the registrant is a shell company (as defined in Rule 12b-2 of the Exchange Act). Yes o No x
The aggregate market value of the registrants common stock held by non-affiliates of the registrant at June 30, 2006, based on the closing price of such stock on the Nasdaq Global Market on such date, was approximately $936 million. The number of shares of the registrants Common Stock, $0.01 par value, outstanding on February 20, 2007, was 68,213,103.
DOCUMENTS INCORPORATED BY REFERENCE
Certain sections of the Proxy Statement to be filed in connection with the 2007 Annual Meeting of Stockholders are incorporated by reference into Part III of this Form 10-K where indicated.
All statements in this Annual Report on Form 10-K that are not historical are forward-looking statements within the meaning of the federal securities laws, including statements regarding our expectations, beliefs, hopes, intentions, strategies or the like. Such statements are based on our current expectations and are subject to a number of factors and uncertainties that could cause actual results to differ materially from those described in the forward-looking statements. We caution investors that there can be no assurance that actual results or business conditions will not differ materially from those projected or suggested in such forward-looking statements as a result of various factors, including, but not limited to, the risk factors discussed in Risk Factors contained in Item 1A of this report and elsewhere in this report. We expressly disclaim any obligation or undertaking to release publicly any updates or revisions to any forward-looking statements contained herein to reflect any change in our expectations with regard thereto or any change in events, conditions, or circumstances on which any such statements are based.
We are engaged in the development, manufacture, sale and service of consumables and systems for genetic analysis in the life sciences and clinical healthcare markets and are recognized as a market leader in creating breakthrough tools that are advancing our understanding of the molecular basis of life. The markets for our products currently include all aspects of molecular biology research in the life sciences, including basic human disease research, genetic analysis, pharmaceutical drug discovery and development, pharmacogenomics (research relating to how a persons genes affect the bodys response to drug treatments), toxicogenomics (research relating to the measurement of gene expression as a predictor of toxicity) and molecular diagnostics. Our integrated GeneChip® microarray platform includes: disposable DNA probe arrays (chips) consisting of nucleic acid sequences set out in an ordered, high density pattern, certain reagents for use with the probe arrays, a scanner and other instruments used to process the probe arrays, and software to analyze and manage genomic or genetic information obtained from the probe arrays. Related microarray technology also offered by us includes licenses for fabricating, scanning, collecting and analyzing results from complementary technologies.
Our business strategy is to capitalize on our leadership position in the DNA microarray field by marketing our GeneChip® technologies to customers based on two central applications: gene expression monitoring and DNA variation detection. Due to the novel, massively parallel approach to studying biological systems that GeneChip® technology enables, numerous discoveries across many disciplines have already been made, as evidenced by the over 7,000 peer-reviewed publications that have cited GeneChip® technology. The application of GeneChip® technologies for diagnosing and guiding treatment of disease is an emerging market opportunity that seeks to improve the effectiveness of health care by collecting information about DNA variation and RNA expression in patients at various times from screening and diagnosis through prognosis and throughout therapeutic monitoring. We currently sell our products directly to pharmaceutical, biotechnology, agrichemical, diagnostics and consumer products companies as well as academic research centers, government research laboratories, private foundation laboratories and clinical reference laboratories in North America and Europe. We also sell our products through life science supply specialists acting as authorized distributors in Latin America, India, the Middle East and Asia Pacific regions, including China.
In March 1992, Affymetrix, Inc. was incorporated in California as a wholly-owned subsidiary of Affymax N.V. (Affymax) and we have continued our business and operations as Affymetrix. We completed our initial public offering in June 1996 and in September 1998 we reincorporated as a Delaware corporation. Our headquarters and principal research and development facilities are located in Santa Clara, California, and we maintain facilities in West Sacramento, California (probe array manufacturing), Sunnyvale, California (administration and array research and development), Emeryville, California (bioinformatics and software development), South San Francisco, California (manufacturing and research and development), Bedford, Massachusetts (instrument development and manufacturing), Singapore (probe array manufacturing, administration, and sales), China (representative office), and additional sales offices in the United Kingdom, Singapore, and Japan.
Introduction to the Genome and its Opportunity
The genetic content of an organism is known as its genome. All known genomes are composed of either deoxyribonucleic acid (DNA) or ribonucleic acid (RNA). The instructions required for every living cell to develop its characteristic form and function are believed to be represented within discrete regions of the DNA or RNA known as genes. The instructions contained within genes are embodied in the specific sequences of the four nucleotide basesadenine-A, cytosine-C, guanine-G and thymine-T (uracil-U replaces T in RNA)that are the chemical building blocks of DNA and RNA. In protein coding genes, the sequence of these building blocks forms a code which instructs the cell to build a protein, comprised of a string of amino acids, ordered in a way which matches the sequence code of the gene. These proteins are an example of a hard copy output of the genetic code and contribute to the structure, biochemical functions and communication mechanisms of the cell in which they are formed.
The DNA molecule possesses a chemical structure which consists of a combination of two DNA strands with hydrogen bonds between nucleotide bases on one strand to complementary nucleotide bases on the other strand. Only certain pairs of the bases can form these complementary bonds: C pairs with G, and A pairs with T. Therefore, a single DNA strand containing bases in the sequence CGTACGGAT can form a bond with a DNA strand containing bases in the sequence GCATGCCTA. Such paired DNA strands are said to be complementary and can form a double helix structure in a process called hybridization. Our GeneChip® technology uses the principle of hybridization to recognize the presence of specific gene sequences and to analyze genetic information.
Genes are segments of DNA that serve as information packets of the genome. In general, a genes functional information is made available to a cell through the process of transcription or gene expression, whereby the sequence is copied into an RNA molecule. Protein coding genes may span thousands to hundreds of thousands, or even millions, of nucleotide bases since the non-coding regions of a gene (called introns) and the coding regions of a gene (called exons) are usually distributed within neighboring genomic sequences that are not translated into proteins or used, or to the extent currently understood, as a functional part of the gene. The number of distinct protein coding genes in the human genome is estimated to be between 25,000 to 30,000. The number of functional non-coding sequences is the focus of current research interest. Though currently unknown, the number of functional non-coding sequences is estimated to be significantly larger than the number of protein coding genes in the human genome.
A primary goal in life sciences research and modern molecular medicine is to unravel the complexities of the genome. This has generated a worldwide effort to identify and sequence the genomes of many organisms. In the human genome, this effort includes more than three billion nucleotide pairs. In recent years, the effort led by the Human Genome Project and related academic, government and industry research projects resulted in a first near complete draft of the human genome sequence. It is anticipated
that many years of research will be required to gain a better understanding of the complexities of the genome, and its characteristics in normal and diseased conditions. We believe that this will lead to a new healthcare paradigm where disease is understood at the molecular level, allowing patients to be diagnosed according to genetic information and then treated with drugs designed to work on specific molecular targets. Ultimately, in addition to diagnosis and treatment, prevention and cure of disease might also be possible based on genetic information.
While scientists are learning more and more about the functions of genes and their variability, there is a great deal more to discover. We believe that the efforts of science to understand the complexities of gene expression, the interaction of genes with our environment and the role of genes in disease and health will continue to provide growth opportunities for our existing gene expression and DNA analysis products, and will continue to create new opportunities in clinical medicine. Toward this end we have partnered with the National Cancer Institute to assemble the first complete map of the human transcriptome (a catalog of all of the RNA transcripts made by the genome). This ongoing effort has already led to the discovery of many novel protein coding and non-protein coding sequences that we have started to include in some of our new products. This effort is also prompting continued development of our sample preparation, array, instrumentation and data analysis technologies.
Genetic Variability and Disease
For the most part, each cell in a complex organism contains a complete copy of the genome. In a population of organisms, individuals vary from one another because of differences in gene sequences which are inherited from each parent and sometimes through the introduction of sequence changes due to environmental damage or biological errors in processes like gene replication. In some cases these variations, or polymorphisms, have little detectable effect on the biology of the organism, while in other cases they may result in an altered biological response to the environment which could thereby lead to disease. By screening for these polymorphisms, researchers seek to identify those that might be implicated in specific diseases. Sometimes it is not a single variation, but the combination of these sequence differences that leads to a diseased state. For this reason, researchers look at the patterns of these polymorphisms in a large number of healthy and affected organisms in order to correlate specific gene polymorphisms with specific diseases.
Another major mechanism by which the fate and function of cells is regulated is the timing and level of gene expression, which can reflect the interface between genes and the environment. Although most cells contain an organisms full set of genes, each cell expresses only a fraction of this set of genes in different quantities and at different times. The expression patterns of genes can be correlated with many human diseases such as cancer, as well as with the effectiveness of treatment in specific patient populations for which new therapies can be developed. By identifying genes that are differentially expressed in particular diseases or patient populations, novel molecular targets and treatments may be identified and validated. In addition, gene expression signatures may be identified that allow the selection of optimal treatment for a single individual.
In order to understand the impact of genomics on health, disease and other aspects of the human condition, scientists must compare both the sequence variation and the gene expression patterns of healthy and diseased individuals, tissues and cells. We believe that our GeneChip® platform not only enables scientists to attain ambitious goals, from identifying genetic variations associated with disease to discovering new drug targets, but also simplifies, accelerates and reduces the cost of understanding this genetic information.
GeneChip® Probe Array Technology
Our GeneChip® technology leverages semiconductor-based photolithographic fabrication techniques, which enables us to synthesize a large variety of predetermined DNA sequences simultaneously in predetermined locations on a small glass chip called a probe array. Photolithography is a technique which uses light to create exposure patterns on the glass chip and direct chemical reactions. The process begins by coating the chip with light-sensitive chemical compounds that prevent chemical coupling. These light-sensitive compounds are called protecting groups. Lithographic masks, which consist of predetermined transparent patterns etched into a glass plate that either block or transmit light, are used to selectively illuminate the glass surface of the chip. Only those areas exposed to light are deprotected, and thus activated for chemical coupling through removal of the light-sensitive protecting groups. The entire surface is then flooded with a solution containing the first in a series of DNA building blocks (A, C, G or T). Coupling only occurs in those regions that have been deprotected through illumination. The new DNA building block also bears a light-sensitive protecting group so that the cycle can be repeated.
This process of exposure to light and subsequent chemical coupling can be repeated many times on the same chip in order to generate a complex array of DNA sequences of defined length. The intricate illumination patterns allow us to build high-density arrays of many diverse DNA sequences in a small area. Unlike conventional synthesis techniques, which generally use a linear process to create compounds, our synthesis technique is combinatorial, in that the number of different compounds synthesized grows exponentially with the number of cycles in the synthesis. Currently available commercial arrays contain over 6.0 million unique sequences. Each unique sequence is 25 nucleotides in length and is represented millions of times within a specified area of the probe array. Just as in the semiconductor industry, we manufacture probe arrays in a wafer format. Each wafer is approximately five inches square and can contain over 300 million unique probe sequences based on current technology. Whole wafers have been used by a related party of Affymetrix, Perlegen Sciences, Inc., in its work to resequence multiple samples of the human genome. For our commercial array products, we can manufacture a large number of identical or different DNA probe arrays on a glass wafer, which is then diced into individual chips. Given the large amount of unique sequences represented in our probe arrays, our technology enables the efficient analysis of a multitude of DNA probes to analyze DNA or RNA sequences in a test sample.
In the semiconductor industry, the principle that the number of transistors in a semiconductor chip doubles every 18 months based on feature shrink, or increased resolution, is known as Moores Law. Because we leverage photolithographic manufacturing processes adapted from the semiconductor industry, we have also been able to continually shrink the size of features, or oligonucleotide probes of a given sequence, on our GeneChip® arrays. Our first commercial GeneChip® products, which we shipped in 1994, had a feature size of 100 microns and by 2005, we introduced our Human Exon Array product with a 5 micron feature size. We have thus been able to increase the amount of genetic information packaged onto our GeneChip® arrays by nearly 400 times since the introduction of our first products.
Because we manufacture our chips in wafer format, we can vary the number of chips manufactured per wafer. We can therefore manufacture thousands of chips per wafer with low information content and lower cost of goods sold or decrease the number of chips per wafer and increase the information content. We expect that we will continue to benefit from this manufacturing leverage as our technology development activities enable further feature shrink. We believe that our unique manufacturing process is a significant competitive advantage.
Our products form an integral part of our GeneChip® system that is designed for use by pharmaceutical, biotechnology, agrichemical, diagnostics and consumer products companies, as well as academic research centers, private and government research foundations and clinical reference laboratories. The GeneChip® system consists of several integrated components: disposable probe arrays containing genetic information on a chip, reagents for extracting, amplifying and labeling target nucleic acids, a fluidics station for introducing the test sample to the probe arrays, a hybridization oven for optimizing the binding of samples to the probe arrays, a scanner to read the fluorescent image from the probe arrays, and software to analyze and manage the resulting genetic information. The function of each single-stranded sequence on the GeneChip® probe array is to bind to its complementary single strand of DNA or RNA from a biological sample. Each unique sequence feature on the GeneChip® probe array contains multiple copies of the same single strand of DNA. The nucleic acid (DNA or RNA) to be tested is isolated from a sample, such as blood or biopsy tissue, amplified and fluorescently labeled by one of several standard biochemical methods. The test sample is then washed over the probe array, where the now labeled individual nucleic acid sequences that represent the genetic content or expressed genes of the sample hybridize to their complementary sequences bound on the array. When scanned by a laser, which is part of the scanner instrument, the test sample generates a fluorescent signal. The locations where a fluorescent signal is detected by an optical detection system on the scanner instrument correspond to sequences complementary to the test sample. Sequence variation, or the quantification of specific sequences of nucleic acids in the sample, can be determined by detecting the relative strength of these signals since the sequence and position of each complementary DNA probe on the probe array is known. The combination of a particular GeneChip® probe array, together with an optimized set of reagents and a user protocol describing how to carry out the procedure, is referred to as an assay.
We currently market products for two principal applications: monitoring of gene or exon expression levels and investigation of genetic variation (DNA analysis including single nucleotide polymorphism (SNP) genotype analysis and resequencing). Our GeneChip® expression monitoring arrays enable our customers to qualitatively and quantitatively measure gene or exon expression levels in a number of frequently studied organisms. Our catalog GeneChip® expression arrays are available for the study of human, rat, mouse and a broad range of other mammalian and model organisms. Human, mouse and rat exon analysis arrays are also available. Additionally, we market CustomExpress, CustomSeq and NimbleExpress products, which enable our customers to design their own custom GeneChip® expression arrays or sequence arrays for organisms of interest to them. Our GeneChip® DNA analysis arrays and variant detection systems are available to enable researchers to perform high throughput polymorphism analysis and to carry out large scale resequencing (comparing the DNA sequence of multiple samples against a known reference sequence, e.g. the published human genome sequence). With its unique, parallel analysis capability, GeneChip® technology enables our customers to perform accurate and cost-effective genetic analysis, using minute amounts of sample DNA, in their own laboratories on a scale that was previously only possible in specialist high throughput centers.
In addition, we believe that genetic analysis and testing products will be a core component in the area of clinical research and molecular diagnostic applications. We are developing our GeneChip® system for clinical research and diagnostic analysis of both gene expression and DNA analysis. The GeneChip® System 3000Dx (GCS3000Dx) is the first microarray instrumentation system for molecular diagnostic laboratories. The GCS3000Dx is 510(k) cleared and CE marked for in vitro diagnostic use in conjunction with the Roche Diagnostics AmpliChip CYP450 Test. The AmpliChip CYP450 Test is the first microarray based diagnostic test and the first product to be released through the Powered by Affymetrix Program.
Together with our collaborative partners, we are focusing on the development and commercialization of diagnostic products in cancer, osteoporosis, cardiovascular, inflammatory, metabolic, infectious and other diseases, and believe that our GeneChip® assays will facilitate more efficient and effective disease detection, prognosis and treatment selection, leading to overall improved patient management. To further our clinical research and molecular diagnostics strategy, we have established partnerships and customer relationships with leading academic researchers, pharmaceutical and biotechnology companies, including F. Hoffmann-La Roche Ltd. (Roche), bioMérieux, Inc. (bioMérieux), and Veridex, LLC, a Johnson & Johnson company (Veridex). We believe that the rapid growth of the clinical research and the diagnostic devices markets holds the potential for GeneChip® technology applications ranging from basic research to clinical trials and, ultimately, diagnostic devices. As a result we are working with leaders in molecular diagnostics to provide custom made GeneChip® probe arrays to their specifications. Our partners subsequently package the chips into kits, seek regulatory approval for their diagnostic use, and sell them into the diagnostic markets using their sales channels. We are leveraging our partners strengths in research, development, regulatory practices and distribution while leveraging our strengths in array technology. These products are marketed as being Powered by Affymetrix.
Gene Expression Monitoring Arrays
Gene expression monitoring is a valuable tool for identifying correlations between genes, determining their biological functions and identifying patterns that might be useful in classifying diseases. To monitor gene expression, we design and manufacture probe arrays with single-stranded DNA molecules that are complementary to sequences within genes or exons of interest. By synthesizing specific probes for multiple genes or exons on a single probe array, we enable researchers to quickly, quantitatively and simultaneously monitor the expression of a large number of genes or exons of interest. By monitoring the expression of such genes under different conditions and at different times, researchers can use the probe arrays to understand the dynamic relationship between gene expression and biological activity. We believe such information will be an important tool in understanding gene function and for the development of new drugs and diagnostic tools. Increasingly, clinical research is showing that gene expression patterns in tissue samples, particularly those from cancerous tissues, can be used to characterize disease sub-types and hopefully to predict therapeutic responses and likely outcomes.
The range of GeneChip® Expression products is described below:
· Standard Expression Monitoring Arrays. We are currently selling a portfolio of standard expression monitoring GeneChip® arrays. Our current offering of standard arrays includes products that monitor the expression of the majority of full-length and partial gene sequences contained in publicly available sequence databases, which correspond with human, mouse, rat, canine, Drosophila melanogaster (fruit fly), Xenopus laevis (frog), Danio rerio (zebrafish), Saccharomyces cerevisiae (yeast), Escherichia coli (bacteria), Pseudomonas aeruginosa (bacteria), Plasmodium falciparum (malarial parasite), Anopheles (mosquito vector of malaria) and Arabidopsis thaliana (plant) organisms.
· Custom Express Arrays. We have established a GeneChip® CustomExpress Array Program enabling customers to design affordable arrays tailored to their specific research needs. Our CustomExpress arrays allow customers to select probes from any of our probe sequences on our catalog arrays and/or to incorporate probes from their own proprietary gene sequences. These arrays are then produced utilizing the same manufacturing technologies as our other GeneChip® whole genome expression arrays.
· Made-to-Order Arrays. We offer the GeneChip® Made-to-Order Array Program to enable our customers to use arrays from selected custom designs and previous-generation GeneChip® arrays which are no longer available as catalog products.
· GeneChip® Exon Arrays. Our exon arrays combine a novel array design strategy with the highest-density array manufacturing capability to provide for the first time, exon-level expression profiling at the whole-genome scale on a single array. The arrays offer new dimensions for researchers to explore the genome so they can investigate global expression of individual exons to uncover and document novel alternative splicing events. They can also obtain more direct comparative genomic information and combine genome-wide exon-level expression data with sequence information to reveal the phenotypic splicing patterns resulting from genetic variations.
· Whole Genome Tiling Arrays. We offer tiling arrays for the genomes of a number of major organisms including human, mouse, drosophila, C.elegans and yeast. These designs use evenly spaced probes across the non-repetitive portion of the genome and rely only on genomic DNA sequence and not functional annotation for their design. The arrays are used for mapping the entire collection of transcribed elements (including non-coding RNAs that are used for structural and regulatory purposes), identifying protein binding and methylation sites and identification of genomic origins of replication. These products are currently available in either Early Access or Technology Access programs, which allow our customers to obtain pre-released versions of our products prior to full commercialization.
DNA Analysis Arrays
As genes and regulatory regions in the human genome are mapped, identified, and sequenced, the value of understanding the variability of sequences among individuals increases. Researchers seek to determine the normal sequence of the gene, which mutations or polymorphisms exist in a population, and whether these variations correlate with a disease or other aspect of the human condition. Studies of the genetics of complex diseases have historically been challenging due to the high costs of sequencing or genotyping of large numbers of affected and unaffected individuals. Genetic variation also impacts how individuals respond to therapeutics. The study of these effects is known as pharmacogenetics. This is part of the broader field of pharmacogenomics, which seeks to understand how the overall composition and expression of the genome affects therapeutic response, drug efficacy and the incidence of adverse side effects to therapy. We believe pharmacogenomics will become increasingly important both in clinical drug trials and patient care. By using our resequencing and genotyping technologies, we believe that our GeneChip® probe arrays could significantly reduce the cost and time required for high-volume polymorphism analysis, which is currently performed through more labor-intensive techniques.
We have initiated product research and development efforts on several genetic analysis probe arrays and variant detection analysis systems and formed collaborations to accelerate the development of our genotyping products. For additional information concerning these efforts and collaborations see the sections of this Form 10-K entitled Research and Development and Our Collaborative Partners. We currently market the following DNA analysis products:
· SNP 5.0 Array. The new single-chip SNP 5.0 Array was made available as a limited release in December 2006 and was commercially launched in February 2007. The new array features single nucleotide polymorphisms (SNPs) from the original two-chip 500K Array Set, as well as more than 420,000 additional probes designed to measure other genetic differences, such as copy number variation. At the time of the commercial launch of the SNP 5.0 Array, Affymetrix software does not support the analysis of these additional probes. The SNP 5.0 Array gives researchers a significant increase in information above the original 500K Array Set for the same price, while reducing the array processing time. The Affymetrix SNP 5.0 Array was developed in collaboration with the Broad Institute of Harvard and the Massachusetts Institute of Technology to better identify and understand the genetic variations associated with complex diseases such as autism, autoimmunity, bipolar disease, cancer, diabetes and heart disease.
· GeneChip® Human Mapping 500K Set. The Mapping 500K Set is the third product in a family of tools for large scale genome-wide genotyping using a simple, scalable, one-primer assay. The 500K Set genotypes over 500,000 SNPs on a two array set. The 500K Set is enabling novel genetic experiments (called genome-wide association studies) to understand the genes involved in complex disease and drug response using real-world disease populations. Researchers have for many years sought to study major diseases in unrelated populations, but the information density (hundreds of thousands of markers) and experiment size (hundreds or thousands of samples) made these experiments unaffordable. The 500K Set is the first and market leading product in making large scale whole genome analysis cost effective.
· GeneChip® Human Mapping 100K Set and 10K Array. We offer two other products based on the same scalable one primer assay as our 500K product: the 100K Set and the 10K Array. The 100K Set genotypes over 100,000 SNPs on a two array set and the 10K Array genotypes over 10,000 SNPs on a single array. The 100K Set was the first product to enable genome-wide association studies and is also used for chromosomal copy number analysis. The 10K Array is used primarily for linkage analysis, the study of the inheritance of disease in families.
· Targeted Genotyping Products. We offer a variety of targeted genotyping products for the analysis of between 1,500 and 20,000 SNPs per sample, using a highly multiplexed assay for use with GeneChip® Universal Tag Arrays. This innovative assay enables customers to genotype selected SNPs in a single assay with only one primer per SNP. We offer the widest variety of standard panels of pre-selected SNPs for major diseases (e.g., inflammatory disease), model organisms (e.g., mouse), and specific applications (e.g., drug metabolizing enzymes). This assay and the products offered through this product line represent the technology and products acquired through the ParAllele acquisition described elsewhere in this report.
· GeneChip® CustomSeq Resequencing Arrays. Version 2 of this product line was released in 2005, increasing sequence content per array by tenfold. The CustomSeq sequence variation product line now enables customers to sequence up to 300,000 bases on one array in just two days with high accuracy. CustomSeq arrays offer researchers a powerful complementary tool to our whole genome genotyping and whole genome expression products on the same industry-standard GeneChip® platform, and are currently being used in a wide variety of applications ranging from infectious disease research to detailed sequence analysis following up association studies to full mitochondrial genome sequencing.
DNA Analysis Products Powered by Affymetrix
· Roche AmpliChip CYP450. The Powered by Affymetrix program and collaboration with Roche announced in January 2003 resulted in the June 2003 release of the Roche AmpliChip CYP450 microarray. The AmpliChip CYP450 microarray allows for complex sequence information to be analyzed for the purpose of genotyping the CYP2D6 and CYP2C19 genes. Sequence variation in these genes can result in marked differences in the way individuals metabolize, and hence respond to, an estimated 25% of all drugs. In late 2004, Roche obtained in-vitro diagnostic status for the product in the United States and Europe.
· FoodExpert-ID assay. bioMérieux has developed and is commercializing the GeneChip® based FoodExpert-ID assay under an industrial license and supply agreement with Affymetrix. The FoodExpert-ID assay detects and identifies, in parallel, DNA sequences from 40 different commercially relevant animal species (and the classes: mammal, fish, bird) in raw and processed food products and in animal feed samples. The FoodExpert-ID assay offers a reliable and sensitive method of authenticating food and animal feed that can be integrated into routine use. The final report of the FoodExpert-ID assay is a permanent record of the vertebrate animal species
composition of a food or animal feed product verified by a DNA signature and can be considered as a product identity card. We manufacture the GeneChip® array, the most critical component of the FoodExpert-ID assay, and it is marketed by bioMérieux as part of our Powered by Affymetrix business model.
Access Programs for Our GeneChip® Arrays
We offer a variety of sales programs for our gene expression monitoring and DNA analysis arrays, tailored to the needs of industrial, biotech and academic/government customers.
Reagents for Our GeneChip® Systems
We offer target labeling and control reagents for use with GeneChip® expression analysis and GeneChip® DNA analysis arrays.
For our gene expression analysis, we offer the following reagents: 3 In Vitro TranscriptionOne-Cycle Target Labeling and Control Reagents, Two-Cycle Target Labeling and Control Reagents, HT One-Cycle Target Labeling and Controls Kit, for use with the GeneChip® Array Station, and GeneChip® Hybridization, Wash, and Stain Kit; Blood RNA Concentration Kit and Globin-Reduction RNA Controls; Whole Transcript AnalysisFor use with Exon and Tiling ArraysGeneChip® WT Sense Target Labeling and Control Reagents, GeneChip® WT cDNA Synthesis and Amplification Kit, GeneChip® WT Terminal Labeling Kit, GeneChip® WT Double-Stranded cDNA Synthesis Kit, GeneChip® WT Amplified Double-Stranded cDNA Synthesis Kit, GeneChip® WT Double-Stranded DNA Terminal Labeling Kit, and GeneChip® Hybridization, Wash, and Stain Kit; ProkaryoticFor use with Prokaryotic ArraysControl Oligo B2, 3nM, DNA Labeling Reagent, Poly-A RNA Control Kit, and GeneChip® Hybridization, Wash, and Stain Kit.
For our DNA analysis, we offer the following reagents: Control Oligo F1, GeneChip® Application-Specific Fixed Assays, GeneChip® Custom SNP Kits and Resequencing Assay Kit.
Instruments for Our GeneChip® Systems
Our GeneChip® instruments provide a fully integrated system for conducting research using GeneChip® probe arrays. The instrument system consists of four hardware devices, each providing for robust preparation and analysis of samples using GeneChip® arrays. The first device is a hybridization oven to control the timing and temperature required for hybridization of the test sample to the probe array. The second device is a fluidics station that controls exposure of the probe array to solutions containing prepared sample and labeled detection reagents across the probe array. The fluidics station can process four probe arrays simultaneously. The fluidics station protocols conclude with a reagent wash that leaves the labeled, hybridized test sample bound to the probe array.
The third device, a laser scanner, is used after completion of protocols on the fluidics and hybridization stations, at which time the cartridge containing the probe array is placed in the scanner and read. The scanner consists of a laser, high-resolution optics, robotics to position and scan the probe array, a fluorescence detector and an interface to a computer workstation. The labeled material that is bound to the hybridized test sample emits fluorescent signals when exposed to the light from the laser. The locations and intensities of the fluorescent signals are recorded by the scanner and stored in the computer for analysis. The fourth device is an autoloader, which is a 48-array carousel that interfaces with the scanner to allow walk-away automation of the scanning steps, while maintaining the loaded arrays at the optimum storage temperature.
The individual components of our GeneChip® instrument system are described in more detail below.
· GeneChip® Hybridization Oven 640. The GeneChip® Hybridization Oven 640 is used to control the timing and temperature required for hybridization of the test sample to the probe array. The GeneChip® Hybridization Oven 640 holds up to eight probe array cartridge carriers (each with eight cartridge slots) that rotate for controlled hybridization of up to 64 probe arrays. This unit delivers temperature control for consistent performance across all probe array applications.
· Fluidics Station 450. The Fluidics Station 450 is used to control exposure of the probe array to solutions containing prepared sample and labeled detection reagents across the probe array and can independently process four probe arrays simultaneously. The fluidics station protocols conclude with a reagent wash that leaves the labeled, hybridized test sample bound to the probe array. Multiple fluidics stations can be connected to the same computer workstation in order to expedite array processing in high throughput laboratories.
· GeneChip® Scanner 3000. The GeneChip® Scanner 3000 uses proprietary laser scanning technology and high resolution optics to read the fluorescent signal from GeneChip® arrays. The GeneChip® Scanner 3000, developed and manufactured by Affymetrix, represents the next generation in scanner technology. The Flying Objective scanning technology incorporated into this scanner has been adapted to provide faster scanning of GeneChip® probe arrays with a high degree of image uniformity and accuracy. The current version of the GeneChip® Scanner 3000 has been validated to allow imaging of GeneChip® arrays with feature sizes as small as 5 microns and can support multicolor assays. The GeneChip® Scanner 3000 is purchased with a computer workstation loaded with Affymetrix GeneChip® Operating Software (GCOS) (discussed below under Software for Our GeneChip® Systems and Analysis Tools).
· GeneChip® AutoLoader. The GeneChip® AutoLoader, developed and manufactured by Affymetrix, is a 48-array carousel that automatically loads and unloads arrays from the scanner, helping researchers automate their array processing and providing walk-away use in the lab. The GeneChip® Scanner with AutoLoader can load and scan 48 current catalog arrays in 28 hours or less, depending on the feature size and array format that is scanned. Thermostatic control allows the stored arrays to be held in a cooled environment and then warmed to optimum temperature for scanning. The AutoLoader attaches to the top of the scanner, saving valuable bench space, and does not require any additional power source.
· Workstations. We offer workstations to accommodate varied research needs. The Type I GeneChip® Workstation (Windows NT) is configured with system software enabling it to operate the GeneChip® Scanner 3000 and multiple Fluidics Stations. Our Type II GeneChip® Workstation (Windows NT and Windows 2000) can operate independently of the scanner and fluidics station instruments.
· GeneChip® Array Station. Through our collaboration with Caliper Life Sciences, Inc., we launched the Affymetrix GeneChip® Array Station that has the capacity to process hundreds of biological samples per week with minimal human supervision, reducing operating costs and variability in target preparation. The GeneChip® Array Station is designed to perform target preparation for both cartridges and array plates as well as perform the wash and stain function for the array plates. The GeneChip® Array Station adapts the same industry-standard GeneChip® technology to a standard 96-well microtiter plate, and runs on an automated system built with off-the-shelf robotic components. In addition, we launched the GeneChip® HT Scanner that utilizes CCD based technology for imaging of the new array plates. The GeneChip® Array Station automates the most labor intensive steps in GeneChip® probe array processing, dramatically reducing the cost per assay. The decrease in cost and increase in throughput makes the GeneChip®
Array Station well suited for downstream development applications such as compound profiling, molecular toxicology and clinical trials.
Software for Our GeneChip® Systems and Analysis Tools
Our GCOS software is supplied as part of an integrated system and runs on an industry standard PC platform. The fluorescence intensity data captured from the scanner are used in conjunction with computer files containing the probe sequence and location of all the probes on the probe array to determine the expression level of a particular gene or to identify particular DNA sequence variations of the test sample.
Our Data Mining Tool® and GeneChip® Operating Software Server (GCOS Server) software products allow for sophisticated analyses of gene expression results and provide a means of linking and integrating this information with other databases.
Customers may choose operating or other software products provided by third party vendors that have been developed through our OpenSystems program, which includes the provision of a Software Developers Kit to interested commercial and academic parties. Through this program we intend to stimulate a wide range of independent groups to develop tools for use with our platform, further enhancing our customers capability to generate unique biological insights from the high quality data provided by the GeneChip® platform.
Finally, our NetAffx Analysis Center (www.affymetrix.com/analysis/) is our exclusive online informatics resource for our customers and provides streamlined, open access to design information and biological annotations associated with our GeneChip® arrays. It was created to assist genomic researchers with the design and analysis of DNA array based experiments. NetAffx offers researchers a searchable catalog of Affymetrix GeneChip® probe array content, a range of publicly available and Affymetrix generated databases, and links to important third party resources.
Instruments for Use in Molecular Diagnostics
The GeneChip® System 3000Dx (GCS3000Dx) is configured especially for the molecular diagnostic market. In September 2004, the GCS3000Dx received Conformite Europeene (CE) certification, clearing it for use as an in-vitro diagnostic (IVD) product in the European Union. In December 2004, the GCS3000Dx was also cleared by the United States Food and Drug Administration (FDA) as an IVD to be used in conjunction with the Roche Diagnostics AmpliChip CYP450 Test. The AmpliChip CYP450 Test also received CE certification and FDA clearance in 2004 making it the first Powered by Affymetrix IVD test. This test, which utilizes an Affymetrix microarray produced specifically for Roche Diagnostics, analyzes a patients Cytochrome P450 2D6 and 2C19 genotypes to look for variations that can influence drug metabolism. The GCS3000Dx will support all Powered by Affymetrix molecular diagnostic tests. The system includes the GCS3000Dx Scanner with Autoloader Dx, FS450Dx Fluidics Station, and Workstation with GCOS Dx software.
We believe that our GeneChip® technology can be effectively applied to complex molecular diagnostic testing. We have formed collaborations and intend to further partner with, or license technology to, established diagnostic and medical device companies to develop, obtain regulatory approval for, and commercialize probe arrays and instrumentation. We anticipate broader use of probe arrays as components of diagnostic products and clinical research applications. We believe that to support large central laboratories, additional instrumentation and automation will need to be developed to allow for handling the large volume testing of the clinical diagnostic setting. To further our molecular diagnostics strategy, we have established a number of collaborations with leading academic researchers, diagnostic companies, pharmaceutical and biotechnology companies.
For example, we are non-exclusively collaborating with Roche to develop and commercialize GeneChip® diagnostic tests for DNA analysis, genotyping and resequencing applications, as well as for RNA expression analysis, in a broad range of human disease areas. Using our GeneChip®technologies, Roche intends to develop and market diagnostic tests for diseases such as cancer and osteoporosis and cardiovascular, metabolic, infectious and inflammatory diseases.
In oncology, we are non-exclusively collaborating with Veridex, a Johnson & Johnson company, to develop and commercialize GeneChip® diagnostic tests for oncology. Using our GeneChip® technologies, Veridex intends to develop and market tests for cancer.
We have a non-exclusive collaborative development agreement and an associated supply agreement for probe arrays with bioMérieux, Inc. in breast cancer and in bacteriology to identify the species and drug resistance profiles of those bacteria causing human infection. The agreements also allow for non-exclusive development of DNA probe arrays for certain diagnostic viral tests and for the fields of food and industrial testing.
We have also entered into non-exclusive collaborative development and associated supply agreement with a number of other smaller companies in focused areas of cancer and certain other diseases.
We and our Powered by Affymetrix partners believe that developing diagnostic products for cancer and other human diseases will establish new standards for molecular diagnostic testing. Ultimately, these products will allow physicians to better diagnose and treat human disease.
We have also entered into a series of agreements with Beckman Coulter that give it the right to develop probe array-based diagnostic products using some elements of our GeneChip® technology. Under these agreements, we have also agreed to grant Beckman Coulter licenses to commercialize probe arrays manufactured using certain of our technologies other than light-directed synthesis. Under the arrangement, Beckman Coulter would pay us transfer prices and royalties on sales of these products.
For additional information concerning our collaborations, see the section of this Form 10-K entitled Our Collaborative Partners.
Our strategy is to establish the GeneChip® system as the platform of choice for analyzing complex genetic information, to expand the applications of our technology and to acquire access to complementary technologies and resources. Accordingly, we have entered into and intend to enter into additional collaborative agreements to further this strategy. The information below is a selected listing of collaborators with whom we have current agreements, together with the related products and programs and the commencement dates of the most recent agreement with each. The section is organized by reference to the product area that represents the most significant portion of the collaboration; however, the collaboration may also involve other areas of our business and product line.
Gene Expression Monitoring
Qiagen, GmbH. In February 2002, we entered into a three-year supply agreement with Qiagen, GmbH (Qiagen) for Qiagen to supply us with certain nucleic acid purification products for use with our GeneChip® arrays for target labeling in expression analysis and have collaboration and co-marketing agreements in place that support the complementary use of Qiagens sample prep instrumentation and reagents with Affymetrix GeneChip® arrays. The term of this supply agreement has been extended for an additional two-year period.
PreAnalytiX GmbH. In October 2003, we entered into a collaborative agreement with PreAnalytiX GmbH, a joint venture between QIAGEN N.V. and Becton, Dickinson and Company. The goal of the
collaboration will be to develop improved methods for the use of PreAnalytiX technology with GeneChip® expression analysis arrays. By combining PreAnalytiX and Affymetrix technologies, our goal is to develop a complete, standardized process for expression profiling starting from whole blood samples.
Invitrogen Corporation. In November 2003 we signed a collaboration and supply agreement to develop and market a new line of GeneChip® brand expression reagents including two new cDNA Synthesis Kits. Both of the new cDNA Synthesis Kits were developed in collaboration with Invitrogen and contain Invitrogens industry-leading SuperScript reverse transcriptase (RT). These reagents have been optimized for use with Affymetrix GeneChip® technology, offering a standardized target preparation system that is easier to use and will help customers produce more robust and consistent array results. The One-Cycle cDNA Synthesis Kit offers reagents for standard target labeling. This protocol has been used by the majority of Affymetrix customers already, but the new kit provides improved configuration and greater convenience. The Two-Cycle cDNA Synthesis Kit offers customers a streamlined procedure for preparing samples using a small amount of material, such as biopsy or laser capture dissected samples. This agreement was extended for two additional years in May 2005.
Perlegen Sciences, Inc. In March 2001, we contributed to Perlegen the rights to use certain intellectual property with no cost basis and received from Perlegen the rights to use and commercialize certain data generated by Perlegen in the array field. Using access to whole-wafer technology developed by Affymetrix, Perlegen focuses on identifying the millions of genetic variations (known as single nucleotide polymorphisms or SNPs) among individuals, and finding patterns in those variations that might be predictive of disease susceptibility or drug response. In January 2003, we obtained accelerated access to Perlegens SNP database which was used to help develop our Mapping 500K Array Set and will also be used in future products. In addition, our collaborative arrangement with Perlegen provides us with access and commercialization rights to certain whole genome technologies, including 248,000 chip-optimized assays covering the genome that we intend to make available to our customers for use in resequencing and genotyping. For additional information concerning our relationship with Perlegen, including our ownership interest in Perlegen, our collaborative relationship with Perlegen and existing relationships between certain of our directors and officers and Perlegen, see Item 7: Managements Discussion and Analysis of Financial Condition and Results of Operations and Note 12 to our Notes to Consolidated Financial Statements of this Form 10-K.
bioMérieux, Inc. In September 1996, we entered into a collaborative development agreement and associated supply agreement for probe arrays with bioMérieux, Inc. (bioMérieux) to identify the species and drug resistance profiles of bacteria causing human infection. As part of the collaboration, bioMérieux is developing molecular diagnostic tests which incorporate these probe arrays. Under the terms of the agreements, bioMérieux provided research and development support and made payments to us upon achievement of certain milestones. In addition, bioMérieux pays specified prices for the supply of probe arrays and royalties on any resulting product sales. In December 1997 and January 1998, we expanded the collaboration with bioMérieux to include the non-exclusive development of DNA probe arrays for molecular diagnostic tests in the fields of virology and food and industrial testing. In March 2003, the collaboration agreement was amended in order to reinstate bioMérieuxs licenses and in March 2005, the collaboration was further expanded to include breast cancer. bioMérieux has launched the FoodExpertID array under this collaboration. See Item 1, Business: DNA Analysis Products Powered by Affymetrix on this Form 10-K.
F. Hoffmann-La Roche Ltd. In February 1998, we entered into a non-exclusive collaborative development agreement with F. Hoffmann-La Roche Ltd. (Roche) to initially develop human probe
array-based diagnostic products. Under the terms of the agreement the parties are collaborating to develop mutually agreed upon arrays, as well as associated instrumentation, software, and reagents. In January 2003, we expanded our collaboration with Roche by granting Roche access to our GeneChip® technologies to develop and commercialize GeneChip® diagnostic laboratory tests for DNA analysis, genotyping and resequencing applications, as well as for RNA expression analysis, in a broad range of human disease areas. Using our GeneChip® technologies, Roche intends to develop and market diagnostic tests for diseases such as cancer and osteoporosis and cardiovascular, metabolic, infectious and inflammatory diseases. Affymetrix and Roche believe that developing targeted microarray expression profiles for cancer, plus genotyping and resequencing profiles for other diseases will enable the creation and commercialization of novel standardized diagnostic solutions. These solutions ultimately will allow physicians to better diagnose and treat human disease. Under the terms of the collaborative agreement, Roche paid us an access fee of $70 million and the agreement also includes a broad range of other compensation payable by Roche to Affymetrix throughout the life of the agreement based on royalties on sales of diagnostic kits and milestone payments for technical and commercial achievements. As part of the agreement, Affymetrix will manufacture and supply Roche with microarrays and related instrumentation based on Affymetrix GeneChip® platform. In 2003, Roche launched the AmpliChip® CYP450 array product initially for research use only, but in late 2004 obtained CE marking and FDA regulatory approvals of the product for in-vitro diagnostic use.
The parties amended the collaborative development agreement in December 2006. Under the terms of the amendment, Roche is relieved of certain future license installment payments that would have been payable by Roche to Affymetrix under the agreement beginning in 2008, Affymetrix is relieved of certain most favorable terms and conditions obligations to Roche, and Roche has agreed to pay to Affymetrix additional milestone payments related to future commercial achievements. The license agreement is subject to Roches option to terminate on December 31, 2010 or any time on or after December 31, 2015, with one years prior notice. As part of the arrangement between the parties, Affymetrix will continue to manufacture and supply Roche with microarrays and related instrumentation based on Affymetrix GeneChip® platform.
Veridex, LLC. In December 2004, we entered into a non-exclusive collaboration agreement with Veridex LLC, a Johnson & Johnson company, to develop and commercialize GeneChip® based diagnostic products for use in the field of oncology. Veridex was granted long-term and comprehensive access to our GeneChip® technology to create and market in-vitro diagnostics for cancer. The agreement, made under the Powered by Affymetrix program, gives Veridex non-exclusive rights to Affymetrix patented arrays, instrumentation systems and planned improvements to these technologies.
Sysmex Corporation. In December 2006, we entered into two agreements with Sysmex Corporation of Kobe, Japan, to distribute, develop and market in-vitro diagnostic tests. A Clinical Development and Marketing Agreement grants Sysmex exclusive rights to distribute Affymetrix diagnostic microarray products and instrumentation in Japan with the opportunity to expand into certain Asia Pacific countries. In addition, a Powered by AffymetrixTM agreement gives Sysmex non-exclusive access to Affymetrix microarray technology to develop and market in-vitro diagnostic tests on a worldwide basis. Sysmex will carry out all registration, sales and customer support activities for the diagnostic instrumentation, array and reagent products as the Affymetrix distributor for Japan. The Powered by AffymetrixTM agreement enables Sysmex to incorporate Affymetrix patented arrays into its diagnostic products. It is expected that the resulting microarray-based tests will enable clinicians to provide more efficient and complete methods to diagnose, classify and manage these disorders.
Broad Institute of Harvard and the Massachusetts Institute of Technology. In September 2004, we entered into a supply and research collaboration agreement with the Broad Institute and Massachusetts
Institute of Technology. The agreement provides Affymetrix technology on preferential terms to the Broad Institute for applications development in return for the option to make commercial products available from the resulting research. The collaboration is focused on whole genome SNP analysis and high throughput expression analysis. The Affymetrix SNP 5.0 Array was developed in collaboration with the Broad Institute to better identify and understand the genetic variations associated with complex diseases such as autism, autoimmunity, bipolar disease, cancer, diabetes and heart disease.
National Cancer Institute. In January 2001, we entered into a collaboration agreement with the National Cancer Institute on a human transcriptome initiative which seeks to construct maps locating the sites of RNA transcription across the entire human genome using high-density whole-genome tiling arrays interrogating at resolutions and throughput rates never before attempted. The transcriptome is defined as the complete collection of transcribed elements of the genome. In addition to well characterized protein coding transcripts, it also represents non-coding RNAs that are used for structural and gene regulatory purposes. Alterations in the structure or levels of expression of any one of these RNAs or their proteins could contribute to disease. An understanding of the transcriptome may provide valuable insights in the research for novel drugs. We have made the data from this initiative freely available to the public databases located at the National Cancer Institute, UC Santa Cruz, GeneExpress, and Affymetrix public websites. This collaboration was extended during fiscal 2003 through mid 2007 to complete the transcript mapping of the entire human genome. In addition to studies performed to detect sites of transcription our technology has been used to map sites of binding by transcription factor regulatory elements, modication of chromatin proteins and DNA and genomic origins of DNA replication.
Caliper Life Sciences, Inc. In January 2004, we signed a collaboration and supply agreement to develop and provide automated target preparation instruments for the GeneChip® Probe Array system. These new automation systems are expected to cut array processing, reduce variability and labor costs, and enable researchers to industrialize their genomic research. The two companies will develop products that leverage Calipers expertise in high-throughput automation and microfluidics with our expertise in microarray technology and applications. The first products were launched in 2004 and automate GeneChip® microarray target preparation steps including hybridization, washing and staining for expression and DNA analysis. The automated system enables a single operator to run up to 96 RNA samples at a time, compared to the manual rate of 20 to 24 samples. In 2005 we launched the Affymetrix GeneChip® Array Station that has the capacity to process hundreds of biological samples per week with minimal human supervision, reducing operating costs and variability in target preparation. The GeneChip® Array Station is designed to perform target preparation for both cartridges and array plates as well as perform the wash and stain function for the array plates. The GeneChip® Array Station adapts the same industry-standard GeneChip® technology to a standard 96-well microtiter plate, and runs on an automated system built with off-the-shelf robotic components.
CapitalBio Corporation, Beijing China. In December, 2005 we entered into an agreement with CapitalBio Corporation to develop and manufacture certain advanced instrumentation for Affyemtrix microarray processing. This agreement was amended in December 2006 to expand the scope of our original instrumentation Development and Supply Agreement. The new agreement establishes a basis for CapitalBio to complete the development of the instrumentation. As part of the agreement, CapitalBio was also granted limited marketing and distribution rights for the instrumentation and associated consumables within China.
The markets for our products include all aspects of molecular biology research in the life sciences, including basic human disease research, clinical research, pharmaceutical drug discovery and development, including pharmacogenomics and toxicogenomics and agricultural research, amongst others. Our customers include pharmaceutical, biotechnology, agrichemical, diagnostics, industrial and consumer
products companies, as well as academic research centers, laboratories in government agencies, private research foundations and clinical and industrial reference laboratories. The following factors, among others, influence the size and development of our markets:
· the availability of genomic sequence and sequence variation data for the human population and for other organisms;
· technological innovation that increases throughput and lowers the cost of genomic and genetic analysis;
· the development of new computational techniques to handle and analyze large amounts of genomic data;
· the availability of government funding for basic and disease-related research;
· the amount of capital and ongoing expenditures allocated to research and development and outsourced spending by biotechnology, pharmaceutical and diagnostic companies for products and services;
· the application of genomics to new areas including molecular diagnostics, agriculture, human identity and consumer goods; and
· the availability of genetic markers and signatures of diagnostic value.
In North America and major European markets, our GeneChip® products are marketed principally through our own sales and distribution organizations. We own or lease sales and service offices in the United States, Europe, Japan and Singapore. In markets outside of North America and Europe, we sell our GeneChip® products principally through third party distributors, primarily in Mexico, India, the Middle East and Asia Pacific, including China. These distributors are life science supply specialists within their own countries and operate as our sole distributors within a defined country or other geographic area.
For molecular diagnostic and industrial applications market opportunities, we supply our partners with arrays and instruments, which they incorporate into diagnostic products and take on the primary commercialization responsibilities. Current collaborative partners include Roche, bioMérieux, Veridex, a Johnson & Johnson company and Sysmex. For additional information concerning our collaborative partners, see Note 5 of Notes to Consolidated Financial Statements: Collaborative Agreements and Item 1, Business: Our Collaborative Partners on this Form 10-K.
We manufacture our GeneChip® probe arrays, GCS 3000 scanner, fluidics stations, instrument control software and certain reagents in-house and contract with third-party suppliers to manufacture our hybridization oven, GCAS and HT Scanner and certain reagents for our GeneChip® system. Additionally, through our GeneChip-compatible application program, a number of third-party software suppliers develop, market and sell genomic data analysis software that interfaces with data files generated by our GeneChip® system.
Our probe array manufacturing process involves wafer preparation, probe synthesis, dicing of synthesized wafers into chips, assembly of chips and quality control. We have developed software programs that extensively automate the design of photolithographic masks used in probe array manufacturing and that control the probe array manufacturing lines. Glass wafers are prepared for synthesis through the application of chemical coatings. GeneChip® probe arrays are synthesized on the wafers using our proprietary, combinatorial photolithographic process. The completed wafers can then be diced to yield individual probe arrays, which are assembled and packaged for shipment.
We are currently manufacturing GeneChip® probe arrays for sale to customers as well as for internal and collaborative purposes. Probe arrays are manufactured at our dedicated manufacturing facility located in West Sacramento, California and in our new manufacturing facility in Singapore. We also maintain a manufacturing process engineering and development facility in Santa Clara, California, and a manufacturing and development facility in Bedford, Massachusetts to support our instrumentation products. In September 2006, we announced our plan to close our Bedford, Massachusetts based instrumentation manufacturing and development facility. We plan to consolidate the Bedford facilitys instrument manufacturing operations with our probe array manufacturing facility in West Sacramento, California. Our instrumentation development capabilities will be consolidated with our principal research and development facilities located in Santa Clara, California. Implementation of the plan began in the fourth quarter of 2006 and is expected to continue into the first half of 2007 and will eliminate or transfer certain positions. See Note 3 of Notes of the Consolidated Financial Statements on this Form 10-K.
All of our commercial instrument and array manufacturing facilities comply with Good Manufacturing Practices as a subset of the Quality System Regulation (21 CFR 820). In 2006, we added capacity to our West Sacramento facility and brought on line and qualified a dedicated GeneChip® probe array manufacturing facility located in Singapore. This facility is fully operational and expected to ramp to high volume production in 2007. In addition, we have put in place the capability to rapidly add incremental capacity if required by demand. See Item 7. Off-Balance Sheet Arrangements and Aggregate Contractual Obligations for further information.
We perform quality tests on selected probe arrays from each wafer and selected probes on such probe arrays. We largely rely on in-process and internal quality control procedures, including controls on the manufacturing process and sample testing, to verify the correct completion of the manufacturing process. In addition, we and our customers rely on the accuracy of genetic sequence information contained in the public databases upon which our products are based. Our probe array manufacturing process is designed to allow us to meet our performance specifications before arrays are shipped. We have a customer inquiry and complaint process in place and we rely on this process to identify and resolve product performance issues that may arise from time to time.
We perform quality tests on in-house and third party reagents at our Santa Clara facility to verify performance of reagent components and assembled reagent kits. We have qualified second source vendors for labeling reagents and oligonucleotides. We perform manufacturing and quality control testing on certain of our reagents at our South San Francisco facility. We also perform services to run certain of our assays on customer samples at our South San Francisco facility.
Key parts of the GeneChip® product line, such as hybridization ovens are available from single sources. We take such steps as we believe are appropriate to ensure that supplies from these vendors are not materially delayed or interrupted, since any such delays or interruptions could in turn delay our ability to deliver these products to our customers. Likewise, certain raw materials or components used in the synthesis of probe arrays or the assembly of instrumentation are currently available only from a single source or limited sources. We take such steps as we believe are appropriate to ensure that materials and components from these vendors are not materially delayed or interrupted, since any such delays or interruptions could in turn delay our ability to produce probe arrays or other components for our GeneChip® system in a timely fashion, in sufficient quantities or under acceptable terms. Alternative sources of supply may be time consuming and expensive to qualify. In addition, we are dependent on our vendors to provide components of appropriate quality and reliability, and to meet applicable regulatory requirements. Accordingly, we also take what we believe are appropriate measures to prevent the delay or interruption of supplies from these vendors and to ensure the appropriate quality for our customers.
We believe a substantial investment in research and development is essential to a long-term sustainable competitive advantage and critical to expansion into high-value markets such as toxicogenomics, pharmacogenomics, and molecular diagnostic and applied testing applications. Our research and development effort is divided into the major areas of basic research, product research and development, and manufacturing process development. Our research and development expenses for the years ended December 31, 2006, 2005 and 2004, were $86.3 million, $77.4 million and $73.4 million, respectively. In addition, on October 21, 2005, we recorded a charge of approximately $8.3 million to acquired in-process technology related to our acquisition of ParAllele BioScience, Inc. (ParAllele), a provider of comprehensive genetic discovery solutions to the life science research, pharmaceutical and diagnostic sectors.
Basic research efforts are carried out through our Affymetrix Research Laboratories to further advance our GeneChip® platform, develop new concepts that can be rapidly productized, and create innovations that will influence our business model in the future. Our initial focus is on basic technology research including high throughput systems, high resolution chip fabrication and detection, genotyping, and gene expression and analysis of the human transcriptome and of other model organism genomes. We are focusing our genotyping research efforts on the development of new assays principally designed to perform whole genome analysis at various resolutions. We believe that products based on this research will ultimately help researchers understand the molecular biology of genomes, to develop more effective therapeutics and help identify the diagnostic markers and tests useful in molecular diagnostic applications.
Product Research and Development
Our product research and development efforts are focused primarily on expanding the applications of the GeneChip® technology including development of new probe array products, improving the overall performance of GeneChip® assays, increasing the information capacity per probe array and simplifying highly complex assays. Our research and development efforts are intended to continue to develop new products based on information from the human and other model organism genomes as well as new genotyping and DNA analysis products. In addition, we intend to continue developing custom product lines for both expression and DNA analysis so that customers can analyze gene expression or DNA variability for any organism. We plan continued software and instrumentation development efforts to enhance the performance and level of automation of our entire GeneChip® system solution.
Manufacturing Process and Development
We are conducting research aimed at enhancing the manufacturing process currently employed in the production of our GeneChip® probe arrays. This process, which leverages semiconductor photolithographic fabrication techniques, is combinatorial in that the number of different compounds synthesized grows exponentially with the number of cycles in the synthesis. The objective of this research is to allow us to produce arrays with higher information density in the same unit area, similar to advances achieved in the semiconductor industry, which has produced silicon chip capacity closely following Moores Law. Moores Law is the observation that the number of transistors per square inch on a silicon chip had doubled every 18 months since the silicon chip was invented. To date, we have also been able to achieve rapid advances in genetic information content, reducing commercial product feature size from 100 microns in 1994 to 5 microns in 2005 with the introduction of our Human Exon Array product. We are continuing research aimed at using smaller feature technology in commercial products and implementing novel, cost-effective packaging approaches for the small array formats including packaging these into the standard industry microtiter plate format.
We rely on a combination of patent, copyright, and trade secret laws, know-how and licensing opportunities to establish and protect our proprietary technologies and products. Our success depends in part on our ability to obtain patent protection for our products and processes, to preserve our copyrights and trade secrets, to operate without infringing the proprietary rights of third parties and to acquire licenses related to enabling technology or products used with our GeneChip® technology.
We are pursuing a patent strategy designed to facilitate our research and development program and the commercialization of our current and future products. While no one patent is considered essential to our success, we aggressively seek to protect our patent rights as our patent portfolio as a whole is material to the success of the business.
There are a significant number of United States and foreign patents and patent applications in our areas of interest, and we believe that there will continue to be significant litigation in the industry regarding patent and other intellectual property rights. Others have filed, and in the future are likely to file, patent applications that are similar or identical to ours or those of our licensors. It may be necessary for us to enter into litigation to defend against or assert claims of infringement, to enforce patents issued to us, to protect trade secrets or know-how owned by us or to determine the scope and validity of the proprietary rights of others. To determine the priority of inventions, it may be necessary for us to participate in interference proceedings declared by the United States Patent and Trademark Office. Litigation or interference proceedings could result in substantial costs to and distraction from our core business and our efforts in respect to such proceedings may not be successful. For further information, see Item 3: Legal Proceedings.
We also rely upon copyright and trade secrets to protect our confidential and proprietary information. We seek to protect our proprietary technology and processes by confidentiality agreements with our employees and certain consultants and contractors. These agreements may be breached, we may not have adequate remedies for any breach and our trade secrets may otherwise become known or be independently discovered by competitors. To the extent that our employees or our consultants or contractors use intellectual property owned by others in their work for us, disputes may also arise as to the rights in related or resulting know-how and inventions.
We are party to various option, supply and license agreements with third parties which grant us rights to use certain aspects of our technologies. We take such measures as we believe are appropriate to maintain rights to such technology under these agreements. In addition, our academic collaborators have certain rights to publish data and information in which we have rights. There is considerable pressure on academic institutions to publish discoveries in the genetics and genomics fields. We take such steps as we believe are appropriate to ensure that such publication will not adversely affect our ability to obtain patent protection for information in which we may have a commercial interest.
Competition in gene expression monitoring, DNA analysis and molecular diagnostics is intense and is expected to increase. Further, the technologies for monitoring gene expression, discovering and analyzing polymorphisms associated with significant diseases, and approaches for commercializing those discoveries are new and rapidly evolving. Currently, our principal competition comes from existing technologies and other DNA array technologies that are used to perform many of the same functions for which we market our GeneChip® systems.
In the gene expression monitoring and DNA analysis fields, existing competitive technologies include gel-based sequencing using instruments provided by companies such as Applied Biosystems, Inc. (an Applera company), and Beckman Coulter, Inc.. Other companies developing or marketing potentially
competitive DNA array technology include: Agilent Technologies, Inc., Applied Biosystems, Inc., BD Biosciences, Clontech, CombiMatrix Corporation, Digital Gene Technologies, Inc., Illumina, Inc., Invitrogen Corporation (formerly Xeotron, Inc.), Molecular Devices, Inc. (formerly Axon Instruments, Inc.), Nanogen, Inc., NimbleGen Systems, Inc., Sequenom, Inc., Solexa, Inc. (formerly Lynx Therapeutics, Inc.), and Visible Genetics, Inc.. For example, companies such as Agilent Technologies, Inc., Applied Biosystems, Inc., and Illumina, Inc. have products for DNA analysis which are directly competitive with our GeneChip® Mapping Array products. In order to compete against existing and emerging technologies, we will need to be successful in demonstrating to customers that the GeneChip® system provides a competitive advantage.
The market for molecular diagnostic products derived from gene discovery is currently limited and highly competitive, with several large corporations already having significant market share. Established diagnostic companies could compete with us by developing new products. Companies such as Abbott Laboratories, Bayer AG, Beckman Coulter, Becton Dickinson, bioMérieux, Johnson & Johnson and Roche have the strategic commitment to diagnostics, the financial and other resources to invest in new technologies, substantial intellectual property portfolios, substantial experience in new product development, regulatory expertise, manufacturing capabilities and the distribution channels to deliver products to customers. Established diagnostic companies also have an installed base of instruments in several markets, including clinical and reference laboratories, which are not compatible with the GeneChip® system and could slow acceptance of our products. In addition, these companies have formed alliances with genomics companies which provide them access to genetic information that may be incorporated into their diagnostic tests.
Future competition in existing and potential markets will likely come from existing competitors as well as other companies seeking to develop new technologies for sequencing and analyzing genetic information. In addition, pharmaceutical and biotechnology companies have significant needs for genomic information and may choose to develop or acquire competing technologies to meet these needs. We have significantly expanded our network of approved service providers in America, Japan, Europe, and China. While these companies expand the reach of Affymetrix technology and make its analytical power available to a wider base of users they may act as a substitute for outright purchase of instruments and arrays by those end users. In the molecular diagnostic field, competition will likely come from established diagnostic companies, companies developing and marketing DNA probe tests for genetic and other diseases, and other companies conducting research on new technologies to ascertain and analyze genetic information. In addition, we have several other third party licensees that could offer products that compete with our product offerings.
Regulation by governmental authorities in the United States and other countries will likely be a significant factor in the manufacturing, labeling, distribution and marketing of certain products and services that may be developed by us or our collaborative partners. In particular, diagnostic products we are developing with our collaborative partners may require regulatory approval by governmental agencies when distributed outside of the research environment.
Commercially available diagnostic tests are regulated as medical devices and are generally subject to rigorous testing and other approval procedures by the United States Food and Drug Administration (FDA). The FDAs Quality System Regulations also apply in connection with our manufacture of arrays and systems as components for use in diagnostic products developed by our partners. Obtaining these clearances or approvals and the compliance with these regulations require the expenditure of substantial resources over a significant period of time, and there can be no assurance that any clearances or approvals will be granted on a timely basis, if at all. Once granted, a clearance or approval may place substantial restrictions on how the device is marketed or labeled or to whom it may be sold. In addition, various
federal and state statutes and regulations govern or influence the manufacturing, safety, storage of our products and components of our products and our record keeping.
Medical device laws and regulations, including those covering in vitro diagnostic products, are also in effect in the European Union, and many of the countries in which we may do business outside the United States. These range from comprehensive device approval requirements for some or all of our medical device products to requests for product data or self-certifications. We may not be able to obtain regulatory approvals in such countries or may incur significant costs in obtaining or maintaining our non-US regulatory approvals. In addition, the export by us of certain of our products which have not yet been cleared for domestic commercial distribution may be subject to FDA or other export restrictions.
Our diagnostic clinical laboratory will be subject to extensive federal and state regulation including the requirements of the Clinical Laboratory Improvement Act of 1988 (CLIA). CLIA is intended to ensure the quality and reliability of clinical laboratories in the United States by requiring all laboratories to meet specified standards in the areas of personnel qualification, administration, participation in proficiency testing, patient test management, quality control, quality assurance and inspections. There can be no assurance that regulations under and future administrative interpretations of CLIA will not have an adverse impact on the potential market for our diagnostic clinical laboratory.
The design of our products and the potential market for their use may be directly or indirectly affected by U.S. and other government regulations governing reimbursement for clinical testing services. The availability of third-party reimbursement for our products and services may be limited or uncertain, particularly with respect to genetic tests and other clinical applications products.
Third-party payors may deny reimbursement if they determine that a prescribed health care product or service has not received appropriate FDA or other governmental regulatory clearances, is not used in accordance with cost-effective treatment methods as determined by the payor, or is deemed by the third-party payor to be experimental, unnecessary or inappropriate. Furthermore, third-party payors are increasingly challenging the prices charged for health care products and services.
Currently, sales of our products and services are not subject to third-party reimbursement. However, we are currently developing diagnostic and therapeutic products with our collaborative partners which may be subject to reimbursement issues. The ability of our collaborators to commercialize such products may depend, in part, on the extent to which reimbursement for the cost of these products will be available under U.S. and foreign regulations governing reimbursement for clinical testing services by government authorities, private health insurers and other organizations. In addition, we are establishing a diagnostic clinical laboratory which will provide subcontract services to primary clinical laboratories.
In the United States, third-party payor price resistance, the trend towards managed health care and legislative proposals to reform health care or reduce government insurance programs could reduce prices for health care products and services, adversely affect the profits of our customers and collaborative partners and reduce our future royalties and product sales.
We are dedicated to compliance and protection of the environment and individuals. Our operations require the use of hazardous materials (including biological materials) which subject us to a variety of federal, state and local environmental and safety laws and regulations. We believe we are in material compliance with current and applicable laws and regulations. However, some of the regulations under the current regulatory structure allow for strict liability, holding a party potentially liable without regard to fault or negligence. We could be held liable for damages and fines as a result of our, or others, business
operations should contamination of the environment or individual exposure to hazardous substances occur. We cannot predict how changes in these laws or development of new regulations will affect our business operations or the cost of compliance.
As of February 20, 2007, we had 1,128 full-time employees. The employee group includes chemists, engineers, computer scientists, mathematicians and molecular biologists with experience in the diagnostic products, medical products, semiconductor, computer software and electronics industries. None of our employees are represented by a collective bargaining agreement, nor have we experienced work stoppages. We believe that we maintain good relationships with our employees. Our success depends in large part on our ability to attract and retain skilled and experienced employees.
Customer demand for probe arrays and instrumentation systems is typically highest in the fourth quarter of the calendar year as customers spend unused budget allocations before the end of the financial year.
Because most customer orders are shipped in the quarter in which they are received, we believe that backlog at quarter end is typically not a material indicator of future sales. In addition, backlog may not result in sales because of cancellation of orders or other factors. On a few occasions we have experienced, and made public announcements about, short-term increases in backlog as a result of factors such as new product introductions or supply constraints.
We operate in one business segment, for the development, manufacture, and commercialization of systems for genetic analysis in the life sciences and diagnostic industry. Our operations are treated as one segment as we only report operating information on a total enterprise level to our chief operating decision-makers. Further, resource allocations are also made at the enterprise level by our chief operating decision-makers.
Our consolidated product and product related revenue from customers outside of the United States for fiscal years 2006, 2005, and 2004 was $168.0 million, $176.7 million and $159.2 million, or 51.9%, 50.5% and 48.1%, respectively, of our consolidated product and product related revenue. A summary of revenues from external customers attributed to each of our geographic areas for the fiscal years ended December 31, 2006, 2005 and 2004, is included in Note 18 of our Consolidated Financial Statements included in this report.
Our internet address is www.affymetrix.com. Information included in our website is not part of this Form 10-K. We make available free of charge on our website our annual reports on Form 10-K, quarterly reports on Form 10-Q, current reports on Form 8-K, and all amendments to those reports as soon as reasonably practicable after such material is electronically filed with or furnished to the SEC. In addition, copies of our annual reports are available free of charge upon written request. The SEC also maintains an Internet site that contains reports, proxy and information statements, and other information regarding issuers that file electronically with the SEC. The address of that site is http://www.sec.gov.
In evaluating our business, you should carefully consider the following risks, as well as the other information contained in this annual report on Form 10-K. If any of the following risks actually occurs, our business could be harmed.
We incurred losses each year from our inception through the year ended December 31, 2003 and have incurred a loss for the year ended December 31, 2006. As a result, we had an accumulated deficit of $120.4 million at December 31, 2006. We expect to continue experiencing fluctuations in our operating results and cannot assure sustained profitability. Our losses have resulted principally from costs incurred in research and development, manufacturing and from selling, general and administrative costs associated with our operations.
Our ability to generate significant revenues and maintain profitability is dependent in large part on our ability to expand our customer base, increase sales of our current products to existing customers, manage our expense growth, and enter into additional supply, license and collaborative arrangements as well as on our ability and that of our collaborative partners to successfully manufacture and commercialize products incorporating our technologies in new applications and in new markets. If our revenues grow more slowly than we anticipate, or if our operating expenses increase more than we expect or cannot be reduced in the event of lower revenues, our business will be materially and adversely affected.
Our quarterly results have historically fluctuated significantly and may continue to fluctuate unpredictably and any failure to meet financial expectations may disappoint securities analysts or investors and result in a decline in our stock price.
Our revenues and operating results may fluctuate significantly due in part to factors that are beyond our control and which we cannot predict. The timing of our customers orders may fluctuate from quarter to quarter. However, we have historically experienced customer ordering patterns for GeneChip® instrumentation and GeneChip® arrays where the majority of the shipments occur in the last month of the quarter. These ordering patterns may limit managements ability to accurately forecast our future revenues. Additionally, license revenue may also be unpredictable and may fluctuate due to the timing of payments of non-recurring licensing fees. Because our expenses are largely fixed in the short to medium term, any material shortfall in revenues will materially reduce our profitability and may cause us to experience losses. In particular, our revenue growth and profitability depend on sales of our GeneChip® products. Factors that could cause sales for these products to fluctuate include:
· our inability to produce products in sufficient quantities and with appropriate quality;
· the loss of or reduction in orders from key customers;
· the frequency of experiments conducted by our customers;
· our customers inventory of GeneChip® products;
· the receipt of relatively large orders with short lead times; and
· our customers expectations as to how long it takes us to fill future orders.
Some additional factors that could cause our operating results to fluctuate include:
· weakness in the global economy and changing market conditions;
· general economic conditions affecting our target customers;
· changes in the amounts or timing of government funding to companies and research institutions; and
· changes in the attitude of the pharmaceutical industry towards the use of genetic information and genetic testing as a methodology for drug discovery and development.
We cannot forecast with any certainty the impact of these and other factors on our sales and operating results in any future period. Results of operations in any period, therefore, should not be considered indicative of the results to be expected for any future period. Because of this difficulty in predicting future performance, our operating results may fall below the expectations of securities analysts or investors in some future quarter or quarters. Our failure in the past to meet these expectations has adversely affected the market price of our common stock and may continue to do so.
We produce our GeneChip® products in an innovative and complicated manufacturing process which has the potential for significant variability in manufacturing yields. We have encountered and may in the future encounter difficulties in manufacturing our products and, due to the complexity of our products and our manufacturing process, we may not fully understand all of the factors that affect our manufacturing processes or product performance.
Although we have taken steps to increase our manufacturing capacity, including the expansion of our existing West Sacramento, California probe array manufacturing facility and the development of a new probe array manufacturing facility in Singapore, there are uncertainties inherent in the expansion of our manufacturing capabilities. For example, manufacturing and product quality issues may arise at the new Singapore facility or as we increase production rates at our West Sacramento facility and launch new products. In addition, we base our manufacturing capabilities on our forecasted product mix for the quarter. If the actual product mix varies significantly from our forecast, we may not be able to fill some orders during that quarter, which could adversely impact our financial results. Difficulties in meeting customer, collaborator and internal demand could also cause us to lose customers or require us to delay new product introductions, which could in turn result in reduced demand for our products.
We manufacture our GeneChip® instrumentation products at our manufacturing facility in Bedford, Massachusetts. In the third quarter of 2006, we announced that we will be consolidating our Bedford instrumentation manufacturing facility with our probe array manufacturing facility in West Sacramento, California as part of our previously announced efforts to reduce expenses in the general and administrative functions including the rationalization of our facilities. The successful reorganization of our instrumentation manufacturing capabilities to our West Sacramento site will require regulatory approvals, including the registration of our West Sacramento site with the United States FDA, and is expected to be substantially completed by the third quarter of 2007. This transition is subject to risks and uncertainties that may cause interruptions in our ability to supply GeneChip® instrumentation products to our customers, which could result in lost revenue and seriously harm our business, financial condition and results of operations.
We rely on internal quality control procedures to verify our manufacturing processes. Due to the complexity of our products and manufacturing process, however, it is possible that probe arrays that do not meet all of our performance specifications may not be identified before they are shipped. If our products do not consistently meet our customers performance expectations, demand for our products will decline. In addition, we do not maintain any backup manufacturing capabilities for the production of our GeneChip® instruments. Any interruption in our ability to continue operations at our existing manufacturing facilities could delay our ability to develop or sell our products, which could result in lost revenue and seriously harm our business, financial condition and results of operations.
The genetic sequence information upon which we rely to develop and manufacture our products is contained in a variety of public databases throughout the world. These databases are rapidly expanding and evolving. In addition, the accuracy of such databases and resulting genetic research is dependent on various scientific interpretations and it is not expected that global genetic research efforts will result in standardized genetic sequence databases for particular genomes in the near future.
Although we have implemented ongoing internal quality control efforts to help ensure the quality and accuracy of our products, the fundamental nature of our products requires us to rely on genetic sequence databases and scientific interpretations which are continuously evolving. As a result, these variables may cause us to develop and manufacture products that incorporate sequence errors or ambiguities. The magnitude and importance of these errors depends on multiple and complex factors that would be considered in determining the appropriate actions required to remedy any inaccuracies. Our inability to timely deliver acceptable products as a result of these factors would likely adversely affect our relationship with customers, and could have a material adverse effect on our business, financial condition and results of operations.
The RNA/DNA probe array field is undergoing rapid technological changes. New technologies, techniques or products could emerge which might allow the packaging and analysis of genomic information at densities similar to or higher than our microarray technology. Other companies may begin to offer products that are directly competitive with, or are technologically superior, to our products. There can be no guarantee that we will be able to maintain our technological advantages over emerging technologies in the future. Over time, we will need to respond to technological innovation in a rapidly changing industry. In addition, although we believe that we are recognized as a market leader in creating systems for genetic analysis in the life sciences, standardization of tools and systems for genetic research is still ongoing and there can be no assurance that our products will emerge as the standard for genetic research. The emergence of competing technologies and systems as market standards for genetic research may result in our products becoming uncompetitive and could cause our business to suffer.
We compete in markets that are new, intensely competitive, highly fragmented and rapidly changing, and many of our current and potential competitors have significantly greater financial, technical, marketing and other resources than we do. In addition, many current and potential competitors have greater name recognition, more extensive customer bases and access to proprietary genetic content. The continued success of our GeneChip® products will depend on our ability to produce products with smaller feature sizes, our ability to dice the wafer, and create greater information capacity at our current or lower costs. The successful development, manufacture and introduction of our new products is a complicated process and depends on our ability to manufacture enough products in sufficient quantity and at acceptable cost in order to meet customer demand. If we fail to keep pace with emerging technologies or are unable to develop, manufacture and introduce new products, including the development of new assay protocols and training our customers, we will become uncompetitive, our pricing and margins will decline and our business will suffer.
Our failure to successfully manage the transition between our older products and our new products may adversely affect our financial results. As we introduce new or enhanced products, we must successfully
manage the transition from older products to minimize disruption in customers ordering patterns, avoid excessive levels of older product inventories and provide sufficient supplies of new products to meet customer demands. When we introduce new or enhanced products, we face numerous risks relating to product transitions, including the inability to accurately forecast demand and difficulties in managing different sales and support requirements due to the type or complexity of the new products.
To be successful, we must attract and retain qualified scientific, engineering, manufacturing, sales, and management personnel. To expand our research, product development and sales efforts we need additional people skilled in areas such as bioinformatics, organic chemistry, information services, regulatory affairs, manufacturing, sales, marketing and technical support. Competition for these people is intense. We will not be able to expand our business if we are unable to hire, train and retain a sufficient number of qualified employees. There can be no assurance that we will be successful in hiring or retaining qualified personnel and our failure to do so could have a material adverse impact on our business, financial condition and results of operations.
We also rely on our scientific advisors and consultants to assist us in formulating our research, development and commercialization strategy. All of these individuals are engaged by other employers and have commitments to other entities that may limit their availability to us. A scientific advisors other obligations may prevent him or her from assisting us in developing our technical and business strategies.
Future competition will likely come from existing competitors as well as other companies seeking to develop new technologies for analyzing genetic information. For example, companies such as Agilent Technologies, Applied Biosystems, and Illumina have products for genetic analysis which are directly competitive with our GeneChip® products. In addition, pharmaceutical and biotechnology companies have significant needs for genomic information and may choose to develop or acquire competing technologies to meet these needs. In the molecular diagnostics field, competition will likely come from established diagnostic companies, companies developing and marketing DNA probe tests for genetic and other diseases and other companies conducting research on new technologies to ascertain and analyze genetic information. Further, in the event that we develop new technology and products that compete with existing technology and products of well established companies, there can be no guarantee that the marketplace will readily adopt any such new technology and products that we may introduce in the future.
The market for molecular diagnostics products is currently limited and highly competitive, with several large companies already having significant market share. Companies such as Abbott Laboratories, Bayer AG, Beckman Coulter, Becton Dickinson, bioMérieux, Celera Diagnostics, Johnson & Johnson and Roche Diagnostics have made strategic commitments to diagnostics, have financial and other resources to invest in new technologies, and have substantial intellectual property portfolios, substantial experience in new product development, regulatory expertise, manufacturing capabilities and the distribution channels to deliver products to customers. Established diagnostic companies also have an installed base of instruments in several markets, including clinical and reference laboratories, which are not compatible with the GeneChip® system and could deter acceptance of our products. In addition, these companies have formed alliances with genomics companies which provide them access to genetic information that may be incorporated into their diagnostic tests.
There have been new accounting pronouncements that may have an impact on our future financial position and results of operations. For instance, in July 2006, the FASB issued Interpretation No. 48,
Accounting for Uncertainty in Income Taxes, an interpretation of FASB Statement No. 109 (FIN 48). Under FIN 48 a company would recognize the benefit from a tax position only if it is more-likely-than-not that the position would be sustained upon audit based solely on the technical merits of the tax position. FIN 48 clarifies how a company would measure the income tax benefits from the tax positions that are recognized, provides guidance as to the timing of the derecognition of previously recognized tax benefits and describes the methods for classifying and disclosing the liabilities within the financial statements for any unrecognized tax benefits. FIN 48 also addresses when a company should record interest and penalties related to tax positions and how the interest and penalties may be classified within the income statement and presented in the balance sheet. FIN 48 is effective for fiscal years beginning after December 15, 2006 and is required to be adopted by us effective January 1, 2007. The cumulative effects, if any, of applying FIN 48 will be recorded as an adjustment to accumulated deficit as of the beginning of the period of adoption. We are currently evaluating FIN 48 and its possible impacts on our financial statements and are not yet in a position to determine such effects, if any.
Our operations are subject to income and transaction taxes in the United States and in multiple foreign jurisdictions. Significant estimates and judgments are required in determining our worldwide provision for income taxes. Some of these estimates are based on interpretations of existing tax laws or regulations. The ultimate amount of tax liability may be uncertain as a result.
Certain jurisdictions have lower tax rates, and the amount of earnings in these jurisdictions may fluctuate. If we do not have profitable operations in these jurisdictions, our tax rate could be adversely impacted. Changes in tax laws and regulatory requirements in the countries in which we operate could have a material impact on our tax provision. To the extent that we are unable to continue to reinvest a substantial portion of our profits in our foreign operations, we may be subject to additional effective income tax rate increases in the future. Tax authorities may challenge the allocation of profits between our subsidiaries and we may not prevail in any such challenge. If we were not to prevail, we could be subject to higher tax rates or double tax.
We rely on our ability to project future taxable income to assess the likelihood that our net deferred tax asset will be realized. To the extent we believe that realization is not more likely than not, we establish a valuation allowance. Significant estimates are required in determining any valuation allowance to be recorded against our net deferred tax assets. Changes in the amount of valuation allowance required may adversely impact our financial results of operations.
Other factors that could affect our effective tax rate include levels of research and development spending and nondeductible expenses such as stock based compensation or merger related expenditures.
We expect that our revenues in the foreseeable future will be derived primarily from products and services provided to a relatively small number of pharmaceutical and biotechnology companies and academic, governmental and other research institutions. Our success will depend upon their demand for and use of our products and services. Our operating results may fluctuate substantially due to reductions and delays in research and development expenditures by these customers. For example, reductions in capital expenditures by these customers may result in lower than expected instrumentation sales and similarly, reductions in operating expenditures by these customers could result in lower than expected GeneChip® array and reagent sales. These reductions and delays may result from factors that are not within our control, such as:
· changes in economic conditions;
· changes in government programs that provide funding to companies and research institutions;
· changes in the regulatory environment affecting life sciences companies and life sciences research;
· market-driven pressures on companies to consolidate and reduce costs; and
· other factors affecting research and development spending.
Our success in penetrating emerging market opportunities in molecular diagnostics depends on the efforts of our partners and the ability of our GeneChip® technologies to be used in clinical applications for diagnosing and informing the treatment of disease.
The clinical applications of GeneChip® technologies for diagnosing and informing the treatment of disease is an emerging market opportunity in molecular diagnostics that seeks to improve the effectiveness of health care by collecting information about DNA variation and RNA expression in patients at various times from diagnosis through prognosis and on to the end of therapy. However, there can be no assurances that molecular diagnostic markets will develop as quickly as we expect or reach what we believe is their full potential. Although we believe that there will be clinical applications of our GeneChip® technologies that will be utilized for diagnosing and informing the treatment of disease, there can be no certainty of the technical or commercial success our technologies will achieve in such markets.
The molecular diagnostics market is relatively new for us and presents us with new risks and uncertainties. Our success in this area depends to a large extent on our collaborative relationships and the ability of our collaborative partners to successfully market and sell products using our GeneChip® technologies. As a result, we are also dependent on the ability of our collaborative partners to achieve regulatory approval for such products in the United States and in overseas markets. Although Roche received FDA approval of the first diagnostic genotyping test for use with our GeneChip® System 3000Dx in late 2004, there can be no assurance that other products using our GeneChip® technologies will achieve needed approvals.
The FDA must approve certain in-vitro diagnostic products before they can be marketed in the U.S. Certain in-vitro diagnostic products must also be approved by the regulatory agencies of foreign governments or jurisdictions before the product can be sold outside the U.S. Commercialization of our and our collaborative partners in-vitro diagnostic products outside of the research environment that may depend upon successful completion of clinical trials. Clinical development is a long, expensive and uncertain process and we do not know whether we, or any of our collaborative partners, will be permitted to undertake clinical trials of any potential in-vitro diagnostic products. It may take us or our collaborative partners many years to complete any such testing, and failure can occur at any stage. Delays or rejections of potential products may be encountered based on changes in regulatory policy for product approval during the period of product development and regulatory agency review. Moreover, if and when our projects reach clinical trials, we or our collaborative partners may decide to discontinue development of any or all of these projects at any time for commercial, scientific or other reasons. Any of the foregoing matters could have a material adverse effect on our business, financial condition and results of operations.
Even where a product is exempted from FDA clearance or approval, the FDA may impose restrictions as to the types of customers to which we can market and sell our products. Such restrictions may materially and adversely affect our business, financial condition and results of operations.
Medical device laws and regulations are also in effect in many countries, ranging from comprehensive device approval requirements to requests for product data or certifications. The number and scope of these requirements are increasing. We may not be able to obtain regulatory approvals in such countries or may
incur significant costs in obtaining or maintaining our foreign regulatory approvals. In addition, the export by us of certain of our products which have not yet been cleared for domestic commercial distribution may be subject to FDA or other export restrictions.
Our diagnostic clinical laboratory is subject to extensive federal and state regulation including the requirements of the Clinical Laboratory Improvement Act of 1988 (CLIA). CLIA is intended to ensure the quality and reliability of clinical laboratories in the United States by requiring all laboratories to meet specified standards in the areas of personnel qualification, administration, participation in proficiency testing, patient test management, quality control, quality assurance and inspections. There can be no assurance that regulations under and future administrative interpretations of CLIA will not have an adverse impact on the potential market for our diagnostic clinical laboratory.
We are currently developing diagnostic and therapeutic products with our collaborators. The ability of our collaborators to commercialize such products may depend, in part, on the extent to which reimbursement for the cost of these products will be available under U.S. and foreign regulations that govern reimbursement for clinical testing services by government authorities, private health insurers and other organizations. In the U.S., third-party payor price resistance, the trend towards managed health care and legislative proposals to reform health care or reduce government insurance programs could reduce prices for health care products and services, adversely affect the profits of our customers and collaborative partners and reduce our future royalties.
We depend on our vendors to provide components of our products in required volumes, at appropriate quality and reliability levels, and in compliance with regulatory requirements. Key parts of the GeneChip® product line are currently available only from a single source or limited sources. In addition, components of our manufacturing equipment and certain raw materials used in the synthesis of probe arrays are available from limited sources. If supplies from these vendors were delayed or interrupted for any reason, we would not be able to get manufacturing equipment, produce probe arrays, or sell scanners or other components for our GeneChip® products in a timely fashion or in sufficient quantities or under acceptable terms. Furthermore, our business is dependent on our ability to forecast the needs for components and products in the GeneChip® product line and our suppliers ability to deliver such components and products in time to meet critical manufacturing and product release schedules. Our business could be adversely affected, for example, if suppliers fail to meet product release schedules, if we experience supply constraints, if we fail to negotiate favorable pricing or if we experience any other interruption or delay in the supply chain which interferes with our ability to manufacture our products or manage our inventory levels.
Maintaining a strong patent position is critical to our business. Litigation on these matters has been prevalent in our industry and we expect that this will continue. Patent law relating to the scope of claims in the technology fields in which we operate is still evolving and the extent of future protection is highly uncertain, so there can be no assurance that the patent rights that we have or may obtain will be valuable. Others have filed, and in the future are likely to file, patent applications that are similar or identical to ours or those of our licensors. To determine the priority of inventions, we will have to participate in interference proceedings declared by the United States Patent and Trademark Office that could result in substantial
costs in legal fees and could substantially affect the scope of our patent protection. We cannot assure investors that any such patent applications will not have priority over our patent applications. Also, our intellectual property may be subject to significant administrative and litigation proceedings such as opposition proceedings against our patents in Europe, Japan and other jurisdictions. Because of the size and breadth of our patent portfolio we may or may not chose to pursue litigation or interferences against those that have infringed on our patents, or used them without authorization, due to the associated expense and time commitment of monitoring these activities. In addition, we have incurred and may in future periods incur substantial costs in litigation to defend against patent suits brought by third parties or initiated by us. For example, we currently are engaged in litigation regarding intellectual property rights with Enzo Life Sciences, Inc. For additional information concerning intellectual property litigation and administrative proceedings, see Part I, Item 3: Legal Proceedings of this Form 10-K. If we fail to protect or to enforce our intellectual property rights successfully, our competitive position could suffer, which could harm our results of operations.
In addition to patent protection, we also rely upon copyright and trade secret protection for our confidential and proprietary information. There can be no assurance, however, that such measures will provide adequate protection for our copyrights, trade secrets or other proprietary information. In addition, there can be no assurance that trade secrets and other proprietary information will not be disclosed, or that others will not independently develop substantially equivalent proprietary information and techniques or otherwise gain access to or disclose our trade secrets and other proprietary information. There can also be no assurance that we will be able to effectively protect our copyrights, trade secrets or other proprietary information. If we cannot obtain, maintain or enforce intellectual property rights, our competitors may be able to offer probe array systems similar to our GeneChip® technology.
Our success also depends in part on us neither infringing patents or other proprietary rights of third parties nor breaching any licenses that may relate to our technologies and products. We are aware of third-party patents that may relate to our technology, including reagents used in probe array synthesis and in probe array assays, probe array scanners, synthesis techniques, polynucleotide amplification techniques, assays, and probe arrays. We routinely receive notices claiming infringement from third parties as well as invitations to take licenses under third party patents. There can be no assurance that we will not infringe on these patents or other patents or proprietary rights or that we would be able to obtain a license to such patents or proprietary rights on commercially acceptable terms, if at all.
We believe that our success in penetrating our target markets depends in part on our ability to develop and maintain collaborative relationships with key companies as well as with key academic researchers. Currently, our significant collaborative partners include Qiagen GmBH for sample preparation and purification systems, Invitrogen Corporation for reagents, and Ingenuity Systems, Inc. for analytical software. We collaborate with both Beckman Coulter Inc. and Caliper Life Sciences in the development of automation for GeneChip® technology applications for use in drug discovery, drug development and clinical research. Roche, bioMérieux and Veridex are collaborative partners in the development of chip products for medical diagnostic and applied testing markets. Relying on these or other collaborative relationships is risky to our future success because:
· our partners may develop technologies or components competitive with our GeneChip® products;
· our existing collaborations may preclude us from entering into additional future arrangements;
· our partners may not obtain regulatory approvals necessary to continue the collaborations in a timely manner;
· some of our agreements may terminate prematurely due to disagreements between us and our partners;
· our partners may not devote sufficient resources to the development and sale of our products;
· our partners may be unable to provide the resources required for us to progress in the collaboration on a timely basis;
· our collaborations may be unsuccessful; or
· we may not be able to negotiate future collaborative arrangements on acceptable terms.
Although we have invested significant other resources to expand our direct sales force and our technical and support staff, we may not be able to establish a sufficiently sized global sales, marketing or technical support organization to sell, market or support our products globally. To assist our sales and support activities, we have entered into distribution agreements through certain distributors, principally in markets outside of North America and Europe. These and other third parties on whom we rely for sales, marketing and technical support in these geographic areas may decide to develop and sell competitive products or otherwise become our competitors, which could harm our business.
A significant amount of our revenue is currently generated from sales outside the United States. Though such transactions are denominated in both U.S. dollars and foreign currencies, our future revenue, gross margin, expenses and financial condition are still affected by such factors as changes in foreign currency exchange rates; unexpected changes in, or impositions of, legislative or regulatory requirements, including export and trade barriers and taxes; longer payment cycles and greater difficulty in accounts receivable collection. We are also subject to general geopolitical risks in connection with international operations, such as political, social and economic instability, potential hostilities and changes in diplomatic and trade relationships. We cannot assure investors that one or more of the foregoing factors will not have a material adverse effect on our business, financial condition and operating results or require us to modify our current business practices.
The risk of product liability claims is inherent in the testing, manufacturing, marketing and sale of human diagnostic and therapeutic products. We may seek to acquire additional insurance for clinical liability risks. We may not be able to obtain such insurance or general product liability insurance on acceptable terms or in sufficient amounts. A product liability claim or recall could have a serious adverse effect on our business, financial condition and results of operations.
Genetic testing has raised ethical issues regarding privacy and the appropriate uses of the resulting information. For these reasons, governmental authorities may call for limits on or regulation of the use of genetic testing or prohibit testing for genetic predisposition to certain conditions, particularly for those that have no known cure. Similarly, such concerns may lead individuals to refuse to use genetics tests even if permissible. Any of these scenarios could reduce the potential markets for our molecular diagnostic products, which could have a material adverse effect on our business, financial condition and results of operations.
We periodically make strategic equity investments in various publicly traded and non-publicly traded companies with businesses or technologies that may complement our business. The market values of these strategic equity investments may fluctuate due to market conditions and other conditions over which we have no control. Other than temporary declines in the market price and valuations of the securities that we hold in other companies will require us to record losses relative to our ownership interest. This could result in future charges on our earnings and as a result, it is uncertain whether or not we will realize any long term benefits associated with these strategic investments.
We have previously engaged in acquisitions and may do so in the future in order to exploit technology or market opportunities. If we acquire another company, we may not be able to successfully integrate the acquired business into our existing business in a timely and non-disruptive manner, or at all. Furthermore, an acquisition may not produce the revenues, earnings or business synergies that we anticipate. If we fail to integrate the acquired business effectively, commercialize the acquired technology, or if key employees of that business leave, the anticipated benefits of an acquisition would be jeopardized. The time, capital management and other resources spent on an acquisition that fails to meet our expectations could cause our business and financial condition to be materially and adversely affected. For example, on October 21, 2005, we completed the acquisition of ParAllele BioScience, Inc. We may not realize the anticipated benefits of the acquisition because of integration and other challenges which include, among others, unanticipated incompatibility of corporate and administrative infrastructures, the ability to retain key employees, including key scientists, the diversion of managements attention from ongoing business concerns, and the potential adverse reaction of customers to an acquisition or resulting changes to the combined companys product and services offerings. In addition, acquisitions can involve substantial charges of stock based compensation that could adversely affect our results of operations.
The market price of our common stock is volatile. To demonstrate this volatility, during the twelve-month period ending on December 31, 2006, the volume of our common stock traded on any given day ranged from 179,200 to 12,591,300 shares. Moreover, during that period, our common stock traded as low as $17.50 per share and as high as $48.00 per share.
Furthermore, volatility in the stock price of other companies has often led to securities class action litigation against those companies. Any future securities litigation against us could result in substantial costs and divert managements attention and resources, which could seriously harm our business, financial condition and results of operations.
During 2006, we conducted an internal review, performed under the direction of our Audit Committee of the Board of Directors, of our historical stock option granting practices from January 1, 1997 through May 31, 2006. As a result of this review, we restated our consolidated financial statements for the years ended December 31, 2005, 2004 and 2003 to record additional non-cash stock-based compensation expense and the related tax impact resulting from stock options granted during fiscal years 1997 to 1999. We have been, and may in the future be, subject to litigation or other proceedings or actions arising in relation to our historical stock option granting practices and the restatement of our prior period financial statements. For example, three purported derivative lawsuits have been filed against us and several of our current and former officers and directors alleging that the defendants breached their fiduciary duty by backdating stock option grants, as well as violations of federal securities laws in
connection with the dissemination of our financial and proxy statements, violations of Generally Accepted Accounting Principles, violations of Section 162(m) of the Internal Revenue Code and violations of state law including violation of the California Corporations Code. In addition, we have received notice from the Securities and Exchange Commission that it is conducting an informal inquiry into our stock option practices. Litigation and any potential regulatory proceeding or action may be time consuming, expensive and distracting from the conduct of our business. The adverse resolution of any specific lawsuit or any potential regulatory proceeding or action could have a material adverse effect on our business, financial condition and results of operations. Also, as a result of our delayed filing our Form 10-Q for the quarter ended June 30, 2006, we will be ineligible to register our securities on Form S-3 for sale by us or resale by others until we have timely filed all periodic reports under the Securities Exchange Act of 1934 for one year. We may use Form S-1 to raise capital or complete acquisitions, which could increase transaction costs and adversely affect our ability to raise capital or complete acquisitions of other companies during this period.
The following chart indicates the facilities that we own or lease, the location and size of each such facility and their designated use.
Information pertaining to legal proceedings can be found in Item 8. Financial Statements and Supplementary DataNote 13. Commitments and Contingencies to the consolidated financial statements, and is incorporated by reference herein.
No matters were submitted during the fourth quarter of the year ended December 31, 2006.
Our common stock is traded on the Nasdaq Global Market under the symbol of AFFX. The following table sets forth, for the periods indicated, the low and high bid prices per share for our common stock as reported by the Nasdaq Global Market.
As of February 20, 2007, there were approximately 451 holders of record of our common stock, one of which is Cede & Co., a nominee for Depository Trust Company (DTC). All of the shares of common stock held by brokerage firms, banks and other financial institutions as nominees for beneficial owners are deposited into participant accounts at DTC and therefore are considered to be held of record by Cede & Co. as one shareholder.
No dividends have been paid on our common stock. We currently intend to retain all future earnings, if any, for use in our business and do not anticipate paying any cash dividends on our common stock in the foreseeable future.
The graph below compares the cumulative total return* on our common stock for the period commencing on December 31, 2001 and ending December 31, 2006 compared to the CRSP Total Return Index for the Nasdaq National Market (U.S. companies) and the CRSP Total Return Index for the Nasdaq Pharmaceutical Stocks (SIC 283). The stock price performance shown on the graph below is not necessarily indicative of future price performance.
* Assumes $100 invested on December 31, 2001 in our common stock and in each index listed above. The total return for our common stock and the indices used assumes the reinvestment of dividends, even though dividends have never been declared on our common stock.
The information under the caption Performance Graph is not deemed filed with the Securities and Exchange Commission and is not to be incorporated by reference in any filing of Affymetrix under the Securities Act of 1933, as amended, or the Securities Exchange Act of 1934, as amended, whether made before or after the date of this 10-K and irrespective of any general incorporation language in such filings.
The following selected historical consolidated financial information has been derived from our audited consolidated financial statements. The balance sheet data as of December 31, 2006 and 2005 and statements of operations data for each of the three years in the period ended December 31, 2006 are derived from audited consolidated financial statements included in this Form 10-K. You should read this table in conjunction with Item 7, Managements Discussion and Analysis of Financial Condition and Results of Operations, and Item 8, Financial Statements and Supplementary Data.
(1) In the third quarter of 2006, we initiated a restructuring plan (the plan) to better align certain of our expenses with our current business outlook. The primary focus of the plan was in the general and administrative functions and included rationalizing our facilities. In August 2006, we terminated certain employees in the selling, general and administrative functions. Additionally, in September 2006, we announced our plan to close our Bedford, Massachusetts based instrumentation manufacturing and development facility. We plan to consolidate the Bedford facilitys instrument manufacturing operations with our probe array manufacturing facility in West Sacramento, California. Our instrumentation development capabilities will be consolidated with our principal research and development facilities located in Santa Clara, California. Implementation of the plan began in the fourth quarter of 2006 and is expected to continue into the first half of 2007, will eliminate or transfer certain positions. We expect the closure of our Bedford, Massachusetts facility to be substantially complete by the third quarter of 2007.
During the third and fourth quarters of 2006, we recognized approximately $13.5 million of expense which was presented in a single line item labeled restructuring in our Statement of Operations.
(2) On October 21, 2005, we completed the $122.4 million acquisition of ParAllele BioScience, Inc. (ParAllele), a provider of comprehensive genetic discovery solutions to the life science research, pharmaceutical and diagnostic sectors.
(3) In September 1999, we issued $150.0 million principal amount of 5% convertible subordinated notes. In February 2000, we issued $225.0 million principal amount of 4.75% convertible subordinated notes.
In August 2001, we repurchased $5.0 million principal amount of the 4.75% notes for total consideration of $3.3 million. In connection with the transaction, we recorded an extraordinary gain of approximately $1.7 million in fiscal 2001 which was reclassified during fiscal 2002 to interest income and other, net in accordance with Financial Accounting Standard (FAS) 145, Rescission of FASB Statements No. 4, 44, and 64, Amendment of FASB Statement No. 13, and Technical Corrections. In December 2002, we repurchased $1.1 million principal amount of the 4.75% notes for total consideration of approximately $0.9 million. In connection with the transaction we recorded a gain of $0.2 million which is included in interest income and other, net. In the second quarter of 2003, we repurchased $53.4 million principal amount of our 4.75% convertible subordinated notes due in 2007 and $48.0 million principal amount of our 5.0% convertible subordinated notes due in 2006. In connection with these transactions, we recognized a net loss of approximately $1.0 million which was included in interest income and other, net. In December 2003, we issued $120.0 million principal amount of 0.75% senior convertible notes. In January 2004, we completed the redemption of our 5.0% notes ($102.0 million face value). In connection with the redemption, we recorded a charge of $3.2 million to interest expense in the first quarter of 2004, related to the unamortized issuance costs and redemption fee associated with the repurchased 5.0% notes. In February 2004, we also completed the redemption of our 4.75% notes ($165.5 million face value). In connection with the redemption, we recorded a charge of $4.9 million to interest expense related to the unamortized issuance costs and redemption fee associated with the repurchased 4.75% notes.
Included in long-term obligations in 2003 through 2006 is the long term portion of deferred revenue relating primarily to our collaboration agreement with Roche. In January 2003, we expanded our collaboration with Roche. Under the terms of the collaboration agreement, Roche paid us an up-front, nonrefundable license fee of $70.0 million. We are recognizing this amount as a component of product related revenue over the research and product development phase, which is expected to approximate five years.
The following discussion should be read in conjunction with the consolidated financial statements and the related notes that appear elsewhere in this document.
This discussion should be read in conjunction with the other sections of this Annual Report on Form 10-K, including Item 1: Business; Item 1A: Risk Factors; Item 6: Selected Financial Data; and Item 8: Financial Statements and Supplementary Data. The various sections of this discussion contain a number of forward-looking statements, all of which are based on our current expectations and could be affected by the uncertainties and risk factors described throughout this filing. Our actual results may differ materially.
All statements in this Annual Report on Form 10-K that are not historical are forward-looking statements within the meaning of Section 21E of the Securities Exchange Act of 1934 as amended, including statements regarding our expectations, beliefs, hopes, intentions, strategies or the like. Such statements are based on our current expectations and are subject to a number of factors and uncertainties that could cause actual results to differ materially from those described in the forward-looking statements. We caution investors that there can be no assurance that actual results or business conditions will not differ materially from those projected or suggested in such forward-looking statements as a result of various factors, including, but not limited to, the risk factors discussed in this Annual Report on Form 10-K in Item 1A. We expressly disclaim any obligation or undertaking to release publicly any updates or revisions to any forward-looking statements contained herein to reflect any change in our expectations with regard thereto or any change in events, conditions, or circumstances on which any such statements are based.
Restatement of Consolidated Financial Statements
On August 9, 2006, we concluded that our consolidated financial statements for the years ended December 31, 2005, 2004 and 2003 should be restated to record additional non-cash stock-based compensation expense, and the related income tax impact, resulting from stock options granted during fiscal years 1997 to 1999 that were incorrectly accounted for under U.S. generally accepted accounting principles. Our decision to restate our financial statements was based on the results of an internal review of our historical stock option granting practices from January 1, 1997 through May 31, 2006 performed under the direction of the Audit Committee of the Board of Directors. The internal review identified certain errors and documentation lapses but did not find any pattern or practice of inappropriately identifying grant dates with hindsight in order to provide discounted or in-the-money grants.
We concluded that there was a material accounting error as a result of a documentation lapse in July 1999, in which the option grant for an aggregate of 1.99 million shares should have been measured for accounting purposes as of a later date. In light of the restatement required by this matter, we also included in this restatement corrections relating to certain other errors and documentation lapses in the fiscal years 1997 and 1998. Approximately 97% of the charges relating to the restatement arose from the documentation lapse in July 1999.
We have determined that the cumulative, pre-tax, non-cash, stock-based compensation expense resulting from revised measurement dates for options granted between 1997 and 1999 was approximately $21.5 million. Accordingly, we recorded stock-based compensation expense of $0, $0, and $1.7 million for the years ended December 31, 2005, 2004 and 2003, respectively, and $19.8 million for the five years preceding 2003, based on the vesting periods of the respective grants. In addition, we recorded an income tax benefit of $8.3 million in 2005. Principally as a result of cumulative losses incurred through 2004, we recorded a full valuation allowance against all deferred tax assets for the years ended December 31, 2004 and 2003 and consequently, there is no income tax effect of the additional stock-based compensation expense recorded in those years. The cumulative effect of the restatement adjustments on our consolidated balance sheet at December 31, 2005 was an increase in additional paid-in capital of $21.5 million offset by an increase in the accumulated deficit of $13.2 million, which results in a net effect on stockholders equity of $8.3 million. The adjustments increased previously reported basic and diluted net income per common share by $0.13 and $0.12, respectively, for the year ended December 31, 2005 and decreased basic and diluted net income per common share by $0.03 for the year ended December 31, 2003 as reported on our Form 10-K/A for the year ended December 31 2005 and our quarterly report filed on Form 10-Q/A filed on August 30, 2006.
As part of our review, we assessed generally whether there were other matters that should have been corrected in our previously issued consolidated financial statements. Apart from the errors underlying the restatement described above, no other matters have come to our attention that should be adjusted in our previously issued consolidated financial statements.
We are engaged in the development, manufacture, sale and service of consumables and systems for genetic analysis in the life sciences and clinical healthcare markets and are recognized as a market leader in creating breakthrough tools that are advancing our understanding of the molecular basis of life. There are a number of factors that influence the size and development of our industry, including: the availability of genomic sequence data for human and other organisms, technological innovation that increases throughput and lowers the cost of genomic and genetic analysis, the development of new computational techniques to handle and analyze large amounts of genomic data, the availability of government and private funding for basic and disease-related research, the amount of capital and ongoing expenditures allocated to research and development spending by biotechnology, pharmaceutical and diagnostic
companies, the application of genomics to new areas including molecular diagnostics, agriculture, human identity and consumer goods, and the availability of genetic markers and signatures of diagnostic value.
We have established our GeneChip® system as the platform of choice for acquiring, analyzing and managing complex genetic information. Our integrated GeneChip® platform includes disposable DNA probe arrays (chips) consisting of gene sequences set out in an ordered, high density pattern; certain reagents for use with the probe arrays; a scanner and other instruments used to process the probe arrays; and software to analyze and manage genomic information obtained from the probe arrays. We currently sell our products directly to pharmaceutical, biotechnology, agrichemical, diagnostics and consumer products companies as well as academic research centers, government research laboratories, private foundation laboratories and clinical reference laboratories in North America and Europe. We also sell our products through life science supply specialists acting as authorized distributors in Latin America, India, the Middle East and Asia Pacific regions, including China. Although we have achieved profitability for the three previous consecutive fiscal years, we incurred a net loss for fiscal 2006. The following overview describes a few of the key elements of our business strategy and our goals:
Increase top-line revenue growth. We intend to increase our top-line revenue through the successful commercialization of our technologies and expansion of our customer base, including leveraging our technologies into new markets. We intend to expand the overall genotyping market opportunity by giving customers the highest possible genetic power for the dollar. We believe that this is a key competitive advantage and will drive commercial momentum by continuing to enable affordable, high-volume genotyping studies. We believe that this continued path toward higher information content and ever more cost-effective products will fuel market growth for years to come. In gene expression, we anticipate expanding the overall market by ensuring that customers have a clear path to clinical studies and the products to get them there. As we continue to enhance the functionality and ease of use of our GeneChip® products, we aim to encourage our customers to expand their uses for our GeneChip® system, which we believe will create new market opportunities for us.
Improve operating efficiency. We intend to leverage our expanded manufacturing capacity and decrease unit costs to improve our product gross margins. We will monitor our expenses closely to ensure our research and development and selling, general and administrative structures remain in line with our revenue growth. In July 2006, we announced our intent to restructure our general and administrative cost structure, including rationalizing our facilities requirements in order to bring our operating expenses in line with our current business requirements. We began those efforts and have incurred charges of approximately $13.5 million in 2006 and saw a decrease in headcount related expenses in our general and administrative functions beginning in the fourth quarter of 2006. We expect to incur additional charges through the third quarter of fiscal 2007 of approximately $14.5 million related to this effort.
On October 21, 2005, Affymetrix acquired 100% of the outstanding shares of ParAllele BioScience, Inc. (ParAllele), a provider of comprehensive genetic discovery solutions to the life science research, pharmaceutical and diagnostic sectors. ParAlleles products and services utilize a unique approach that leverages novel biochemical processes rather than complex instrumentation to discover and analyze minute variations in the human genome. We expect the acquisition to accelerate the development and commercialization of new products and create greater opportunities for market penetration and revenue generation as well as increase our core assay development capabilities.
The results of operations of ParAllele have been included in the accompanying consolidated financial statements from the date of acquisition. See Note 4 for further information regarding this acquisition.
CRITICAL ACCOUNTING POLICIES & ESTIMATES
The following section of Managements Discussion and Analysis of Financial Condition and Results of Operations is based upon our consolidated financial statements, which have been prepared in accordance with U.S. generally accepted accounting principles. The preparation of these financial statements requires management to make estimates and assumptions that affect the reported amounts of assets, liabilities, revenue and expenses, and related disclosure of contingent assets and liabilities. Management bases its estimates on historical experience and on various other assumptions that are believed to be reasonable under the circumstances, the results of which form the basis for making judgments about the carrying values of assets and liabilities that are not readily apparent from other sources. Actual results may differ from these estimates under different assumptions or conditions.
Our significant accounting policies are fully described in Note 2 to our consolidated financial statements. However, certain accounting policies are particularly important to the reporting of our financial position and results of operations and require the application of significant judgment by our management. An accounting policy is deemed to be critical if it requires an accounting estimate to be made based on assumptions about matters that are highly uncertain at the time the estimate is made, and if different estimates that reasonably could have been used, or changes in the accounting estimates that are reasonably likely to occur periodically, could materially impact the financial statements. Management believes the following critical accounting policies reflect its more significant estimates and assumptions used in the preparation of the Consolidated Financial Statements.
We enter into contracts to sell our products and, while the majority of our sales agreements contain standard terms and conditions, there are agreements that contain multiple elements or non-standard terms and conditions. As a result, significant contract interpretation is sometimes required to determine the appropriate accounting, including whether the deliverables specified in a multiple element arrangement should be treated as separate units of accounting for revenue recognition purposes, and if so, how the value of the arrangement should be allocated among the deliverable elements, when and how to recognize revenue for each element, and the period over which revenue should be recognized. We recognize revenue for delivered elements only when the fair values of undelivered elements are known and customer acceptance, if required, has occurred. Changes in the allocation of the sales price between delivered to undelivered elements might impact the timing of revenue recognition, but would not change the total revenue recognized on any arrangement.
We evaluate the collectibility of our trade receivables based on a combination of factors. We regularly analyze our significant customer accounts, and, when we become aware of a specific customers inability to meet its financial obligations to us, such as in the case of bankruptcy filings or deterioration in the customers operating results or financial position, we record specific bad debt allowances to reduce the related receivable to the amount we reasonably believe is collectible. We also record allowances for bad debt on a small portion of all other customer balances based on a variety of factors including the length of time the receivables are past due, the financial health of the customer, macroeconomic considerations and historical experience. If circumstances related to specific customers change, our estimates of the recoverability of receivables could be further adjusted.
We enter into inventory purchases and commitments so that we can meet future shipment schedules based on forecasted demand for our products. The business environment in which we operate is subject to rapid changes in technology and customer demand. We perform a detailed assessment of inventory each period, which includes a review of, among other factors, demand requirements, product life cycle and development plans, component cost trends, product pricing, product expiration and quality issues. Based on this analysis, we record adjustments to inventory for potentially excess, obsolete or impaired goods, when appropriate, in order to report inventory at net realizable value. Revisions to our inventory adjustments may be required if actual demand, component costs, supplier arrangements, or product life cycles differ from our estimates.
NON-MARKETABLE EQUITY SECURITIES
As part of our strategic efforts to gain access to potential new products and technologies, we invest in equity securities of certain private biotechnology companies. Our non-marketable equity securities are carried at cost unless we determine that an impairment that is other than temporary has occurred, in which case we write the investment down to its impaired value. We periodically review our investments for impairment; however, the impairment analysis requires significant judgment in identifying events or circumstances that would likely have significant adverse effect on the fair value of the investment. The analysis may include assessment of the investees (i) revenue and earnings trend, (ii) business outlook for its products and technologies, (iii) liquidity position and the rate at which it is using its cash, and (iv) likelihood of obtaining subsequent rounds of financing. If an investee obtains additional funding at a valuation lower than our carrying value, we presume that the investment is other than temporarily impaired. We have experienced impairments in our portfolio due to the decline in equity markets over the past few years.
Income tax expense is based on pretax financial accounting income. Under the liability method, deferred tax assets and liabilities are determined based on the difference between the financial statement and tax basis of assets and liabilities using enacted tax rates in effect for the year in which the differences are expected to reverse. We must then assess the likelihood that the resulting deferred tax assets will be realized. To the extent we believe that realization is not more likely than not, we establish a valuation allowance. Significant estimates are required in determining our provision for income taxes, our deferred tax assets and liabilities, and any valuation allowance to be recorded against our net deferred tax asset. Some of these estimates are based on interpretations of existing tax laws or regulations. We believe that our estimates are reasonable and that our reserves for income tax related uncertainties are adequate. Various internal and external factors may have favorable or unfavorable effects on our future effective tax rate. These factors include, but are not limited to, changes in tax laws, regulations and/or rates, changing interpretations of existing tax laws or regulations, changes in the valuation of our deferred tax assets or liabilities, future levels of research and development spending, nondeductible expenses, changes in overall levels and geographical mix of pretax earnings and ultimate outcomes of income tax audits.
We are subject to legal proceedings principally related to intellectual property matters. Based on the information available at the balance sheet dates, we assess the likelihood of any adverse judgments or outcomes to these matters, as well as potential ranges of probable losses. If losses are probable and reasonably estimable, we will record a reserve in accordance with SFAS 5, Accounting for Contingencies. Any reserves recorded may change in the future due to new developments in each matter.
ACCOUNTING FOR STOCK-BASED COMPENSATION
Beginning January 1, 2006, we account for employee stock-based compensation in accordance with FASB Statement No. 123R, Share-Based Payment (SFAS 123R). Under the provisions of SFAS 123R, we estimate the fair value of our employee stock awards at the date of grant using the Black-Scholes option-pricing model, which requires the use of certain subjective assumptions. The most significant of these assumptions are our estimates of the expected term, volatility and forfeiture rates of the awards. The expected stock price volatility assumption was determined using a combination of historical and implied volatility of our common stock. We determined that blended volatility is more reflective of market conditions and a better indicator of expected volatility than historical volatility. The estimate of these key assumptions is based on historical information and judgment regarding market factors and trends. As required under the accounting rules, we review our valuation assumptions at each grant date and, as a result, we are likely to change our valuation assumptions used to value employee stock-based awards granted in future periods.
SFAS 123R requires that employee stock-based compensation costs be recognized over the requisite service period, or the vesting period, in a manner similar to all other forms of compensation paid to employees. Accordingly, in 2006, we recognized employee stock-based compensation of $1.4 million in cost of product sales, $3.7 million in research and development expense and $10.0 million in selling, general and administrative expenses, which includes approximately $1.1 million related to the modification of an executive officers stock option grant to increase the associated exercise period. We also incurred a charge of $7.5 million which was included in restructuring related to the modification of a research and development employees equity awards, all of which became fully vested under the terms of a prior severance agreement when he was involuntarily terminated as part of our cost reduction efforts. We adopted SFAS 123R on a modified prospective basis. As of December 31, 2006, $20.4 million of total unrecognized compensation cost related to non-vested stock options not yet recognized is expected to be allocated to cost of products sales and operating expenses over a weighted-average period of 1.6 years.
There was no stock-based compensation expense related to employee stock options recognized under SFAS 123R during the fiscal years prior to 2006.
We have recently engaged in, and may continue to engage in, restructuring actions, which require management to utilize significant estimates related to expenses for severance and other employee separation costs, lease cancellation, realizable values of assets that may become duplicative or obsolete, and other exit costs. If the actual amounts differ from our estimates, the amount of the restructuring charges could be materially impacted. For a full description of our restructuring actions, see Note 3.
In July 2006, the FASB issued Interpretation No. 48, Accounting for Uncertainty in Income Taxes, an interpretation of FASB Statement No. 109 (FIN 48). Under FIN 48 a company would recognize the benefit from a tax position only if it is more-likely-than-not that the position would be sustained upon audit based solely on the technical merits of the tax position. FIN 48 clarifies how a company would measure the income tax benefits from the tax positions that are recognized, provides guidance as to the timing of the derecognition of previously recognized tax benefits and describes the methods for classifying and disclosing the liabilities within the financial statements for any unrecognized tax benefits. FIN 48 also addresses when a company should record interest and penalties related to tax positions and how the interest and penalties may be classified within the income statement and presented in the balance sheet. FIN 48 is effective for fiscal years beginning after December 15, 2006 and is required to be adopted by us effective January 1, 2007. The cumulative effects, if any, of applying FIN 48 will be recorded as an adjustment to accumulated
deficit as of the beginning of the period of adoption. We are currently evaluating FIN 48 and its possible impacts on our financial statements and are not yet in a position to determine such effects, if any.
The following discussion compares the historical results of operations for the years ended December 31, 2006, 2005 and 2004.
The components of product sales are as follows (in thousands, except percentage amounts):
Total product sales decreased in 2006 as compared to 2005. Probe array sales declined primarily due to a decrease of $25.4 million in revenue related to the reduction of the average selling price of our probe arrays primarily due to a decrease in the selling price of our current two-chip 500K genotyping product and a decrease of $2.2 million in revenue due to a decline in unit sales. Instrument sales decreased primarily due to a decrease of $9.7 million in revenue related to a decline in unit sales of our GeneChip® Scanner 3000, a decrease of $2.3 million due to a drop in unit sales of our GeneChip® Scanner 3000 high-resolution upgrades and a decrease of $3.2 million in unit sales of our automation equipment.
In July 2006, we announced a decrease in the selling price of our current two-chip 500K genotyping product as we believe this will enable scientists to run more samples, and thereby increase the power of their genetic association studies. We believe this decrease in the selling price could cause our future probe array revenues to decline from prior periods if unit sales do not increase.
Total product sales increased in 2005 as compared to 2004. Probe array sales increased primarily due to an increase of $69.3 million in revenue related to growth in unit sales of our GeneChip® probe arrays. This increase was partially offset by a decrease of $39.6 million in revenue related to a decline in the average selling price of our arrays due to a change in product mix. Reagent sales increased primarily due to an increase in the average selling price of our reagents related to a change in product mix and growth in reagent unit sales associated with increased probe array unit sales. Instrument sales decreased primarily due to a decrease of $16.4 million in revenue related to a decline in unit sales of our GeneChip® Scanner 3000 upgrades as we end our GeneArray® Scanner 2500 upgrade cycle as well as a decrease of $4.4 million in revenue related to a decline in unit sales and in the average selling price of our Fluidics Station 450. These decreases were partially offset by growth of $4.4 million in revenue related to increased unit sales of our new automation equipment and growth of $4.1 million in revenue related to increased unit sales of our GeneChip® Scanner 3000 high-resolution upgrades.
The components of product related revenue are as follows (in thousands, except percentage amounts):
Total product related revenue increased in 2006 as compared to 2005 primarily due to an increase in service and other revenue of approximately $19.3 million in revenue related to our genotyping services business. This increase was partially offset by the decline in subscription fees earned under our GeneChip® array access programs as we continue to transition customers to volume-based pricing on array sales.
In January 2003, under the terms of an expanded collaborative agreement, Roche paid us an access fee of $70 million, which we are recognizing as a component of product related revenue in license fees over the estimated research and development period of approximately five years. The expanded collaboration agreement includes a broad range of other compensation payable by Roche to Affymetrix throughout the life of the agreement based on royalties on sales of diagnostic kits and milestone payments for technical and commercial achievements.
The parties amended the collaborative development agreement in December 2006. Under the terms of the amendment, Roche is relieved of certain future license installment payments that would have been payable by Roche to Affymetrix under the agreement beginning in 2008, Affymetrix is relieved of certain most favorable terms and conditions obligations to Roche, and Roche has agreed to pay to Affymetrix additional milestone payments related to future commercial achievements. The license agreement is subject to Roches option to terminate on December 31, 2010 or any time on or after December 31, 2015, with one years prior notice. As part of the arrangement between the parties, Affymetrix will continue to manufacture and supply Roche with microarrays and related instrumentation based on Affymetrix GeneChip® platform.
Total product related revenue decreased in 2005 as compared to 2004. The decrease was primarily due to the decline in subscription fees earned under our GeneChip® array access programs as we continue to transition customers to volume-based pricing on array sales. Additionally, the recognition of $2.2 million in milestone revenue due to the achievement of substantive goals as defined by our collaborative agreements in 2004 contributed to the decrease. Service and other revenue increased in 2005 primarily due to an increase of approximately $3.2 million in other revenue related to DNA analysis services pertaining to the completion of certain contracts assumed by us upon the acquisition of ParAllele in the fourth quarter of 2005 and due to increased sales of service contracts for our GeneChip® Scanner 3000.
The increase in royalties and other revenue in 2006 as compared to 2005 was primarily related to the recognition of $9.5 million related to a license agreement entered into in the second half of 2006 with no continuing performance obligations. We also recorded an additional $5.3 million in deferred revenue related to this license agreement which will be recognized when the payment is received which we anticipate will occur during 2008.
Royalties and other revenue decreased in 2005 as compared to 2004 primarily due to a decline in the number of our non-core license agreements and the recognition of one monetary and one non-monetary license agreement with no continuing performance obligations in 2004 partially offset by the recognition of a $1.5 million license agreement with no continuing performance obligations in the fourth quarter of 2005.
Perlegen revenue increased in 2006 as compared to 2005 primarily related to increased demand by Perlegen for our commercial probe arrays. The increase of $11.2 million was partially offset by a decrease in demand for wafers used by Perlegen for internal research and development of $3.1 million. Probe arrays and wafers used by Perlegen for use in their internal research and development activities are sold at our fully burdened cost of manufacturing. We also earn a royalty on certain of these wafers if they are used by Perlegen for certain revenue activities. Royalties recognized were $0.9 million and $2.8 million in 2006 and 2005, respectively. We rely on reports from Perlegen to recognize royalty revenue.
Perlegen revenue increased in 2005 as compared to 2004 primarily due to increased demand by Perlegen for our probe arrays and whole wafers. Probe arrays and wafers used by Perlegen for use in their research and development activities are sold at our fully burdened cost of manufacturing. We also earn a royalty on certain of these wafers if these products are used by Perlegen for certain revenue activities. Royalties recognized were $2.8 million and $1.4 million in 2005 and 2004. Beginning in the fourth quarter of 2005, we also began selling probe arrays and reagents to Perlegen for use in certain new applications, with 2005 revenues totaling $1.0 million.
The 8.3 point decrease in product gross margin in 2006 as compared to 2005 is primarily due to the following three factors: 3.2 points of the decrease can be attributed to a decrease in the average selling price of our probe arrays primarily due to a decrease in the selling price of our two-chip 500K genotyping product; 3.5 points of the decrease is due to underutilized capacity in our probe array and instrument manufacturing facilities and increased manufacturing costs for our chips; 1.5 points of the decrease is due to start-up costs for our new Singapore manufacturing facility. These decreases were partially offset by a 1.6 point increase in product gross margins as we sold more of our higher margin products.
In July 2006, we announced a decrease in the selling price of our current two-chip 500K genotyping product as we believe that this will enable scientists to run more samples, and thereby increase the power of their genetic association studies.
The 21.7 point decrease in product related gross margin in 2006 as compared to 2005 was primarily due to an increase in expenses related to our genotyping services business as well as the continuing decline in sales of our access agreements as we continue to transition customers to volume-based discounting on product sales.
The gross margin percentage on product sales increased in 2005 as compared to 2004. Of the 1.5 point increase in product gross margins, approximately 1.0 point of the increase can be attributed to an increase in probe array gross margins due to the conversion of a royalty bearing technology license from Oxford Gene Technology (OGT) to a fully paid-up non-royalty bearing license in the second half of 2004. The remainder of the gross margin increase can be attributed to decreased unit costs associated with the increased utilization of our probe array manufacturing facility due to increased sales volume and a change in our overall product mix towards our higher margin products such as probe arrays and reagents. The increase was partially offset by a decrease in gross margin of approximately 1.0 point which can be attributed to a $2.0 million warranty adjustment and related yield issues associated with the release of our Mapping 500K Array product in the third quarter of 2005.
The gross margin percentage on product related revenue decreased in 2005 as compared to 2004. The margin decreases were primarily due to a decline in sales of our access agreements as we continue to transition customers to volume-based pricing on array sales and the recognition of $2.2 million in milestone revenue due to the achievement of substantive goals as defined by our collaborative agreements in 2004. We also incurred costs related to the completion of contracts entered into by ParAllele in the fourth quarter of 2005.
Cost of Perlegen revenue was flat in 2006 as compared to 2005 primarily due to decreased demand by Perlegen for wafers used for internal research and development which are sold at our fully burdened cost of manufacturing which was offset by an increase in the costs of commercial products purchased by Perlegen.
Cost of Perlegen revenue increased in 2005 and increased as compared to 2004 primarily due to increased demand by Perlegen for our probe arrays and whole wafers.
The increase in research and development expenses in 2006 as compared to 2005 was primarily due to an increase in headcount related costs of $5.6 million primarily due to our acquisition of ParAllele as well as an increase in new hires in the first half of 2006 and an increase of $3.7 million of stock-based compensation expense related to our adoption of SFAS 123R.
The increase in research and development expenses in 2005 as compared to 2004 was primarily due to an increase in the consumption of development supply materials of $2.1 million and an increase of $1.9 million in co-development costs. These increases are as a result of our preparation for new product introductions and applications expected to be introduced in the next twelve months. These increases were partially offset by a decrease of $2.5 million in performance based compensation.
The increase in selling general and administrative expenses in 2006 as compared to 2005 was partially due to $10.0 million of stock-based compensation expense related to our adoption of SFAS 123R. Our stock-based compensation expense included a $1.1 million charge in the second quarter of 2006 due to the modification of an executive officers stock option grant to include an extension of time to exercise as well as the vesting of certain stock options without a requisite service period. Additionally, legal expenses increased by $4.1 million in 2006 as we continue to defend our patent rights. In the third quarter of 2006, we also incurred approximately $1.1 million in costs related to our stock option review. Lastly, we saw an increase related to higher headcount-related costs of $6.3 million in 2006.
We anticipate the cost reduction elements of our restructuring plan implemented in the third quarter of 2006 will decrease our headcount costs in future periods as we continue to bring our general and administrative costs in line with our current business requirements. We expect legal costs to vary substantially as the intensity of legal activity changes. There can be no assurance that we have adequately estimated our exposure for potential damages associated with pending or future litigation.
Selling, general and administrative expenses increased in 2005 as compared to 2004. These increases were primarily due to higher headcount-related costs from hiring new employees and normal salary increases for existing employees of $6.3 million and an increase of $2.5 million in consulting costs. Our increased headcount expenses and consulting costs were primarily due to our continued international expansion, investments in new strategic initiatives, and in anticipation of our new product introductions. These increases were partially offset by a decrease of $6.2 million in performance based compensation.
With our adoption of SFAS 123R on January 1, 2006 all stock-based compensation expense has been included in cost of product sales, research and development expenses, selling, general and administrative expenses and restructuring charges. Under SFAS 123R, stock-based compensation cost is calculated on the date of grant using the Black-Scholes method. The compensation cost is then amortized straight-line over the vesting period.
In 2005, we recorded stock-based compensation expense related to restricted stock awards granted to certain employees and one fully vested restricted stock award granted to a non-employee that were calculated using prior accounting standards. Also in 2005, stock-based compensation included the amortization of deferred stock compensation generated upon the acquisition of ParAllele in October 2005. The amortization of deferred stock-based compensation was principally attributable to research and development and selling, general and administrative employees. In 2004, stock-based compensation related to the amortization of deferred stock compensation generated upon the acquisition of Neomorphic in October 2000.
During the year ended December 31, 2005, we recorded a charge of approximately $8.3 million to acquired in-process technology upon the October 21, 2005 completion of our acquisition of ParAllele BioScience, Inc. (ParAllele), a provider of comprehensive genetic discovery solutions to the life science research, pharmaceutical and diagnostic sectors. We utilized the assistance of a third party valuation specialist to help us determine the estimated fair value of certain research and development programs in-process at the acquisition date that had not yet reached technological feasibility and had no alternative future use. These projects primarily included the development of a DNA genotyping product and to a lesser extent certain RNA and other expression products. The fair values of these projects were determined using the Income Approach whereby we estimated each projects related future net cash flows between 2005 and 2015 and discounted them to their present value using a risk adjusted discount rate of 28%. This discount rate is based on our estimated weighted average cost of capital adjusted upward for the risks associated with the projects acquired. The projected cash flows from the acquired projects were based on estimates of revenues and operating profits related to the projects considering the stage of development of each potential product acquired, the time and resources needed to complete the development and approval of each product, the life of each potential commercialized product and the inherent difficulties and uncertainties in developing products and services based on complex genetic technologies and biochemical processes. As of October 21, 2005, the DNA genotyping project was 33% complete and the RNA & other projects were 40% complete. We expect to incur costs of approximately $5.6 million to complete the DNA genotyping project and costs of $1.5 million to complete the RNA & other projects. As of December 31, 2006, there was no significant update to the percentage of completion of the projects, nor the anticipated costs to complete as certain of the projects were temporarily delayed during 2006 as we re-prioritized our efforts in connection with the restructuring plan. These projects are expected to resume in 2007.
The estimates used by us in valuing the licensed technologies and acquired in process technologies were based upon assumptions we believe to be reasonable but which are inherently uncertain and unpredictable. Our assumptions may be incomplete or inaccurate, and no assurance can be given that unanticipated events and circumstances will not occur. Accordingly, actual results may vary from the projected results.
In the third quarter of 2006, we initiated a restructuring plan (the plan) to better align certain of our expenses with our current business outlook. Our primary focus of the plan was in the general and administrative functions and included rationalizing our facilities. In August 2006, we terminated certain employees in the selling, general and administrative functions. Additionally, in September, we announced our plan to close our Bedford, Massachusetts based instrumentation manufacturing and development facility. We plan to consolidate the Bedford facilitys instrument manufacturing operations with our probe array manufacturing facility in West Sacramento, California. Our instrumentation development capabilities will be consolidated with our principal research and development facilities located in Santa Clara, California. Implementation of the plan began in the fourth quarter of 2006 and is expected to continue into the first half of 2007 and will eliminate or transfer certain positions. We expect the closure of our Bedford, Massachusetts facility to be materially complete by the third quarter of fiscal 2007.
We estimate that the total restructuring expenses to be incurred in connection with the plan will be approximately $28 million. Of this total, we estimate that $19 million relates to employee severance and relocation benefits, $7 million relates to contract termination costs associated with vacating the leased Bedford facility, and approximately $2 million relates to other restructuring costs such as fixed asset write downs and equipment and inventory relocation costs. We have incurred restructuring expenses beginning in the third quarter of 2006 and expect to continue incurring related expenses through the third quarter of 2007. Cash outlays incurred in connection with these restructuring activities are estimated to be approximately $20 million. In accordance with SFAS 146, Accounting for Costs Associated with Exit or Disposal Activities (SFAS 146), the costs relating to employee severance and relocation will be accrued over the remaining service periods of the employees.
During the third and fourth quarters of 2006, we recognized approximately $13.5 million of expense which was presented in a single line item labeled restructuring in our Statement of Operations, of which approximately $7.5 million of the expense related to the modification of a former Neomorphic employees equity awards, all of which became fully vested under the terms of a prior leave of absence agreement when he was involuntarily terminated in connection with the plan.
The components of interest income and other, net, are as follows:
Interest income increased in 2006 as compared to 2005 due to an increase in the returns on our cash and marketable securities balances. In 2006, we also realized a $1.6 million currency gain on assets held in various foreign currencies. In 2005, we realized losses related to the recognition of our proportionate share of Perlegens net loss. The carrying value of our investment in Perlegen was written down to zero at December 31, 2005; therefore, to the extent we do not make any additional future investments in Perlegen, we do not expect further expenses related to this investment.
Interest income and other, net, increased in 2005 as compared to 2004 primarily due to an increase in our cash and marketable securities balances as generated by positive cashflows from operations and financing activities. However, these increases were partially offset by a realized loss related to the recognition of our proportionate share of Perlegens net loss.
Interest expense remained flat in 2006 as compared to 2005. Interest expense in 2006 and 2005 primarily consists of interest and amortization of the debt issuance costs associated with our $120.0 million 0.75% senior convertible notes issued in December 2003. Interest expense was higher in 2004 primarily due to the write off of approximately $3.8 million in issuance costs plus $4.3 million in redemption premiums incurred by us in connection with the redemption of our 5% and 4.75% convertible subordinated notes in January 2004.
The provision for income taxes in 2006 is more than the 35% U.S. federal statutory rate applied to consolidated loss before income taxes primarily due to losses incurred in foreign jurisdictions which do not result in tax benefits and the tax impact of non-deductible stock based compensation expenses under SFAS 123R. The renewal during the fourth quarter of the federal research and development tax credit in the United States reduced our income tax provision by approximately $702K or 12%. In 2005, the benefit for income taxes was more than the 35% U.S. federal statutory rate primarily due to the reduction in the valuation allowance for certain deferred tax assets. In 2004, the provision for income taxes was more than the 35% U.S. federal statutory rate primarily due to the full valuation allowance against net deferred tax assets.
Statement of Financial Accounting Standards No. 109, Accounting for Income Taxes (SFAS 109) provides for the recognition of deferred tax assets if realization of such assets is more likely than not. As of December 31, 2006, current net deferred tax assets totaled $13.5 million. Non-current net deferred tax assets totaled $29.2 million. We considered whether it is more likely than not that some portion or all of the deferred tax assets will not be realized, using historical book income, the scheduled reversal of temporary differences, and projected future book and taxable income in making this assessment.
Based upon the weight of available evidence, as of December 31, 2006, we provided for a valuation allowance of $68.0 million against our net deferred tax assets. $58.7 million of the valuation allowance is recorded to offset benefits relating to stock options. This amount of the valuation allowance will not be
reduced until the benefit reduces tax that would otherwise be due. The remaining $9.3 million is related to acquired company tax benefits which will reduce goodwill when recognized.
If we determine that we would not be able to realize all or part of our net deferred tax assets in the future, we would record a charge to income and an adjustment to the deferred tax assets in the period we make that determination.
As of December 31, 2006, we had net operating loss carryforwards of $110.7 million for U.S. federal purposes, which expire in the years 2018 through 2025. For state purposes, we had net operating loss carryforwards of $52.3 million, which expire in the years 2010 through 2015. Additionally, we have federal and state research and development tax credit carryforwards and other various tax credit carryforwards of approximately $40.5 million, which expire in the years 2008 through 2026 if not utilized. Utilization of our net operating loss and tax credit carryforwards may be subject to substantial annual limitations due to the ownership change limitations provided by the Internal Revenue Code and similar state provisions. Such an annual limitation could result in the expiration of the net operating loss or tax credits before utilization.
Cashflow (in thousands)
Cash provided by operating activities is net (loss) income adjusted for certain non-cash items and changes in operating assets and liabilities. Cash provided by operating activities was comprised of a net loss of $13.7 million, non-cash depreciation and amortization of $22.8 million, non-cash stock-based compensation expense of $15.1 million and restructuring of $13.5 million. Significant changes in assets and liabilities are as follows:
Our total accounts receivable, including Perlegen, was $77.8 million at December 31, 2006, a decrease of $19.3 million from December 31, 2005. The decrease is primarily due to lower product sales of probe arrays and instruments.
Our inventory balance was $46.5 million at December 31, 2006, an increase of $10.5 million from December 31, 2005. The increase is primarily due to the buildup of consumables as we continue to transition production to our Singapore facility.
Our accounts payable and accrued liabilities balance was $62.9 million at December 31, 2006, a decrease of $8.7 million from December 31, 2005. The decrease was primarily due to the payment of accounts payable related to our manufacturing expansion in West Sacramento and capital costs on our new manufacturing facility in Singapore.
Our investing activities, other than purchases, sales and maturities of available-for-sale securities, primarily consist of capital expenditures, strategic investments and purchased technology rights.
Cash used for capital expenditures was $79.3 million, $40.2 million and $21.9 million for the years ended December 31, 2006, 2005 and 2004, respectively. Our capital expenditures in 2006 primarily related to our manufacturing expansion in West Sacramento and capital costs related to our new manufacturing facility in Singapore. Capital expenditures in 2004 and 2005 related to continued expansion in our operating facilities, investments in information management systems, and purchases of production and lab equipment.
In 2005, we made a $2.0 million purchase of non-marketable equity investment in Perlegen Sciences. Additionally, we used $10.9 million in cash, net of cash assumed, for our acquisition of ParAllele in October 2005.
In 2004, the primary component of purchased technology rights related to a $41.9 million license payment to OGT.
Our financing activities primarily consist of activity under our employee stock plans. Cash provided by the issuance of stock under our employee stock plan was $9.3 million, $74.7 million and $36.2 million in 2006, 2005 and 2004, respectively. Net cash used in financing activities in 2004 was primarily attributable to the $271.8 million cash outflow for the redemption of our 5% and 4.75% convertible subordinated notes.
Off-Balance Sheet Arrangements and Aggregate Contractual Obligations
As part of our ongoing business, we do not participate in transactions that generate relationships with unconsolidated entities or financial partnerships, such as entities often referred to as structured finance or special purpose entities (SPEs), which would have been established for the purpose of facilitating off-balance sheet arrangements or other contractually narrow or limited purposes. As of December 31, 2006, we are not involved in any SPE transactions.
The impact that our contractual obligations as of December 31, 2006 are expected to have on our liquidity and cash flow in future periods is as follows (in thousands):
(1) Our 0.75% senior convertible notes are due in 2033, however holders may require us to repurchase all or a portion of their notes on December 31, 2008, 2013, 2018, 2023, and 2028.
(2) Purchase commitments include agreements to purchase goods or services that are enforceable and legally binding on Affymetrix and that specify all significant terms, including: fixed or minimum quantities to be purchased; fixed, minimum or variable price provisions; and the approximate timing of the transaction. Purchase obligations exclude agreements that are cancelable without penalty.
(3) Other commitments relate to an ongoing obligation held by the Company to contribute $0.5 million in a venture capital limited partnership. The venture capital limited partnership may call in this amount, in part or in full at any time according to the contract terms. Other commitments also includes a funding obligation to support two fellowships under the Bio-X Program at Stanford.
We have financed our operations primarily through product sales, sales of equity and debt securities, collaborative agreements, interest income, and licensing of our technology. As of December 31, 2006, we had cash, cash equivalents, and available-for-sale securities of approximately $247.8 million. We anticipate that our existing capital resources along with the cash to be generated from operations will enable us to maintain currently planned operations and planned capital expenditures (estimated to be approximately $35 million for the year ending December 31, 2007, which includes the completion of our manufacturing expansion), for the foreseeable future. However, this expectation is based on our current operating, financing and capital expenditure plans, which are subject to change, and therefore we could require additional funding. Our long-term capital expenditure requirements will depend on numerous factors including: the expansion of commercial scale manufacturing capabilities; our ability to maintain existing collaborative and customer arrangements and establish and maintain new collaborative and customer arrangements; the progress of our research and development programs; initiation or expansion of research programs and collaborations; the costs involved in preparing, filing, prosecuting, defending and enforcing intellectual property rights; the effectiveness of product commercialization activities and arrangements; the purchase of patent licenses; and other factors.
As of December 31, 2006, we have no credit facility or other committed sources of capital. To the extent capital resources are insufficient to meet future capital requirements; we will have to raise additional funds to continue the development of our technologies. There can be no assurance that such funds will be available on favorable terms, or at all. To the extent that additional capital is raised through the sale of equity or convertible debt securities, the issuance of such securities could result in dilution to our stockholders. If adequate funds are not available, we may be required to curtail operations significantly or to obtain funds by entering into collaboration agreements on unattractive terms. Our inability to raise capital would have a material adverse effect on our business, financial condition and results of operations.
Our exposure to interest rate risk relates primarily to our investment portfolio. Fixed rate securities may have their fair market value adversely impacted due to fluctuations in interest rates, while floating rate securities may produce less income than expected if interest rates fall. Due in part to these factors, our future investment income may fall short of expectations due to changes in interest rates or we may suffer losses in principal if forced to sell securities which have declined in market value due to changes in interest rates.
The primary objective of our investment activities is to preserve principal while at the same time maximize yields without significantly increasing risk. To achieve this objective, we invest our excess cash in debt instruments of the U.S. Government and its agencies and high-quality corporate issuers, and, by policy, restrict our exposure to any single corporate issuer by imposing concentration limits. To minimize the exposure due to adverse shifts in interest rates, we maintain investments at an average maturity of less than three years.
The table below presents principal amounts and related weighted interest rates by year of maturity for our available-for-sale securities and debt obligations, are primarily denominated in United States dollars, and the fair value of each as of December 31, 2006 and 2005. Our available for sale securities are classified on our balance sheet in cash and cash equivalents, available-for-sale securitiesshort-term and available-for-sale securitieslong-term.