Back to GetFilings.com

SECURITIES AND EXCHANGE COMMISSION
Washington, D.C. 20549

FORM 10-K

For the fiscal year ended December 31, 2000

Commission File Number 000-31147

DELTAGEN, INC.
(Exact name of registrant as specified in its charter)

1003 Hamilton Avenue, Menlo Park, California 94025
(Address of principal executive offices) (Zip Code)
Telephone Number: (650) 752-0200

SECURITIES REGISTERED PURSUANT TO SECTION 12(b) OF THE ACT:

SECURITIES REGISTERED PURSUANT TO SECTION 12(g) OF THE ACT:

Common Stock $0.001 Par Value
(Title of Class)

    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 / /

    Indicate by check mark if disclosure of delinquent filers pursuant to Item 405 of Regulation S-K (§ 229.405 of this chapter) is not contained herein, and will not be contained, to the best of registrant's 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. / /

    At March 1, 2001, the aggregate market value of the registrant's Common Stock held by non-affiliates of the registrant was approximately $75,147,765 .

    At March 19, 2001, the number of shares outstanding of registrant's Common Stock was 29,758,332.

DOCUMENTS INCORPORATED BY REFERENCE

    Portions of Definitive Proxy Statement for the Company's 2001 Annual Meeting of Stockholders to be held on April 26, 2001, are incorporated by reference into Part III of this Form 10-K where indicated.

Deltagen, Inc.
Index to Annual Report on Form 10-K
For the fiscal year ended December 31, 2000

2

PART I

Item 1. Business

Overview

    Deltagen is a leading provider of genomics based information, products and programs that are used to discover drug targets and drug candidates and to accelerate pharmaceutical and biotechnology drug and product development. We research, develop and use technologies that build upon our technology platforms to discover the function, role, usefulness and disease relevance of mammalian genes and secreted proteins in drug discovery and development. Our technology enables us to select and target genes we believe to be relevant to disease and delete, or "knockout," these genes in mice. We then use an extensive, integrated analysis program to assess the function and potential pharmaceutical relevance of these genes and the proteins these genes encode. We believe that our technology allows us to knockout genes in mice and to make discoveries on the mammalian function, role, usefulness and disease relevance of genes and secreted proteins in drug discovery faster and on a larger scale than has been previously possible.

    We are implementing a strategy to integrate our:

•

data and information generated from our knockout animal models;



•

mammalian gene function and secreted protein discoveries;



•

commercialization of the intellectual property we generate on the use of mammalian genes and secreted proteins in drug development through alliances and collaborations with others and our own internal products and programs;



•

generation of information, products and services for pharmaceutical and biotechnology drug discovery efforts; and



•

development and expansion of our own products and programs in collaboration with other companies and through our own internal programs.

    We have established collaborations and relationships with major pharmaceutical and biotechnology companies worldwide to accelerate the discovery of and to commercialize therapeutic and diagnostic products to improve human and animal health. These companies include GlaxoSmithKline plc, Merck & Co., Inc., Pfizer Inc. and Schering-Plough Research Institute.

    We believe that our ability to determine gene function, to develop products and to identify potential drug candidates results from our leveraging of our powerful technology platforms. Our genomics technologies, processes and information systems are fully integrated with one another and rapidly generate information on the function and relationships between genes and the proteins these genes encode and the usefulness of genes as new drug targets and proteins as new drug candidates. We have used these systems to establish our products and programs that include our:

•

large scale program to generate mammalian gene knockout animals and to discover gene function;



•

DeltaBase™ portfolio of gene knockout animal models and mammalian gene function data analysis and management database;



•

DeltaXpress™ microarray-based mammalian gene knockout expression data program;



•

mammalian gene knockout secreted protein discovery programs, including our Delta-GT™ secreted gene trap program; and



•

internal early-stage biopharmaceutical product development programs, including our CD-123 antigen program in-licensed as a potential treatment for Acute Myelogenous Leukemia.

3

Background

Overview

    Pharmaceutical and biotechnology companies are continually challenged to develop and market increased numbers of drugs. This challenge has led to increased research and development spending and the development of a new research focus called genomics-based drug discovery. This new research effort involves understanding the relationship between genes and the functions they regulate. An organism's genetic information, or genome, is comprised of deoxyribonucleic acid, or DNA molecules. DNA itself is comprised of four different chemical subunits called nucleotide bases that are strung together in a precise sequence. Encoded within a DNA sequence are discrete sets of instructions, or genes, that collectively serve to regulate our biological processes by producing proteins. Alterations in gene sequence, or mutations, form the basis of many diseases.

    Understanding the critical role that genes play in regulating biological processes and disease has led to efforts to obtain information on all the genes contained within the human genome and the genomes of other organisms. International public and private genomics projects have generated vast amounts of data and identified all the genes within the human genome. The draft of the complete human genome was released in February 2001. The human genome is comprised of approximately three billion nucleotide bases that encode approximately 30,000 to 40,000 genes. Approximately 3,000 to 10,000 of these genes and the proteins these genes encode may have potential as drug targets and drug candidates. Seeking to capitalize upon the opportunity to discover new drug targets, pharmaceutical, biotechnology and genomic companies are rapidly pursuing genomics-based drug discovery programs. We believe that a system that will enable a more rapid commercialization of these newly discovered genes and the proteins these genes encode can be of significant value to drug manufacturers.

    Genomics-based drug discovery generally consists of:

•

discovering and identifying DNA sequences that make up the genes within the genome;



•

determining the function of the discovered genes so that their role in regulating biological processes and disease can be understood;



•

using information on gene function and disease relevance to assess the value of a particular gene or its protein product as a target for drug discovery; and



•

in the case of genes that are potential drug targets, utilizing high-volume chemistry and other drug discovery methods to target the relevant gene to produce a commercially viable drug.

    Pharmaceutical, biotechnology and academic researchers have performed the initial task of identifying genes. However, identifying the genes is only the first hurdle of several significant current impediments to genomics-based drug discovery. The next key hurdles are determining gene function, identifying which genes can serve as viable drug targets and which proteins encoded by these genes can serve as viable drug candidates. Determining gene function with respect to a biological process or disease is a complex undertaking that requires extensive and detailed physiological analysis.

Discovering Gene Function

    The scientific community has attempted to find efficient methods of determining the functions of individual genes for several decades. This process is particularly challenging for the pharmaceutical industry because drug development requires a very precise understanding of potential drug discovery targets. It is important that a pharmaceutical or biotechnology researcher understands all the possible ramifications of targeting a gene or its associated protein with a drug, including any potentially serious side effects of drug administration.

4

    The drug discovery and development process is an expensive, time consuming and lengthy process. Before a gene can be selected as a candidate for the drug discovery and development process, its complete functional role must be determined as thoroughly as possible. Determining whether a gene is a relevant target for drug discovery is a process termed target validation. Currently, researchers generally use the standard or genetic approaches to drug target discovery and validation described below.

Standard Approach to Target Discovery and Validation

    The objective of the standard approach to target discovery is to sort through the tens of thousands of gene sequences to find ones that can be analyzed using current techniques for determining in vivo biology, or the function of the gene in a living organism. Under the standard approach, researchers:

•

identify a gene sequence;



•

isolate and make an operational copy of the gene in order to facilitate physiological analysis of its function;



•

find the tissues where the gene is active, or expressed, which may provide clues about the potential functional role of the gene;



•

perform cell-based, or in vitro, experiments using potentially relevant cell types to define a potential role for the gene; and



•

conduct studies in a living mammal, or in vivo studies, to determine the role of the gene in a whole organism, a key step in providing confirmation of the gene as a validated target for drug discovery.

    The standard approach to target discovery is a time consuming, expensive and multi-staged process in which only a limited number of genes reach the final steps of the validation process. The lack of in vivo data early in this process can lead to the selection of genes based on criteria that do not necessarily or accurately reflect their functions in a living organism. This can lead to significant wasted time, effort and expense in selecting genes that represent valid targets.

Genetic Approach to Target Discovery and Validation

    Since the function of a gene in an animal can vary widely from its function as determined by in vitro studies, it is preferable to obtain IN VIVO data at an early stage in the drug discovery process. To accomplish this, some pharmaceutical and biotechnology companies have employed a genetic approach that initially uses non-mammalian organisms to determine IN VIVO function. Under the genetic approach, researchers:

•

choose a lower organism, such as a fly or worm, based on the compatibility of the organism with the specific organ system or function to be studied;



•

create a functional mutation in the lower organism by using chemicals to produce a permanent genetic alteration that is reflected by an observable change in the organism;



•

identify the mutated gene responsible for the observed change;



•

find the equivalent gene in mammals; and



•

conduct IN VIVO studies to determine the role of the gene in a whole organism, a key step in providing confirmation of the gene and the protein encoded by that gene as a validated target or candidate for drug discovery.

    The genetic approach to target discovery is subject to a number of limitations. Under the genetic approach, researchers randomly mutate the genome. This may result in the identification of genes with

5

interesting functions; however, these genes may not become valid drug targets because only certain subsets of genes are amenable to current drug discovery methods. In addition, since lower organisms are far less complex than mammals, they do not have many of the mammalian genes and their corresponding physiological functions. Thus, while lower organisms can provide information on gene function similarity with humans, their ability to provide information concerning how genes control mammalian physiology is limited. As a result, validation typically requires mammalian studies which are traditionally time-consuming and costly.

Determining Mammalian Gene Function

    During the past decade, the preferred method for determining a gene's function in mammals has been to disrupt, or knockout, the gene in a mouse and to assess the physiological, pathological and behavioral consequences of removing the gene from the animal. The results of this analysis can determine the function and disease relevance of a particular gene and the potential of the gene and the protein encoded by that gene as a drug target or drug candidate.

    Mice and humans are both higher mammals, and their genomes are similar in size and gene content. Therefore, performing knockouts of genes in mice has advantages over studies in non-mammalian organisms for defining the function and disease relevance of human genes. Additionally, mice are one of the few mammals for which approaches to genetic manipulation have been established. Because of the high degree of physiological and genetic similarity between mice and humans, the mouse gene knockout system has the potential to become an effective and widely accepted model for target validation studies.

    A drawback of this model though has been the low-volume, high-cost and commercially unfeasible time-frames for production. Traditional approaches to create mouse knockouts allow a research team to create only a limited number of knockouts per year. As a result, mouse knockouts have been used as the last step of the target validation process, if at all.

    Despite the time-frame and labor intensive nature of the process, the academic scientific community has adopted the mouse knockout as a model for gene function studies. Information from these studies is often publicly available. However, this information is often fragmentary, difficult to obtain and is selectively and non-uniformly reported. In addition, when such information is available, it can be difficult to cross-reference or compare using standardized medical/scientific vocabulary or to compare with pre-existing models of disease.

    Collectively, these limitations have made mouse knockouts difficult to use as a first-line drug discovery tool despite their utility in determining gene function.

Our Solution

    We have developed a fully integrated target validation system that provides gene function information based on mouse knockouts at early stages of drug target discovery. Our solution moves directly from gene identification to determination of gene function in a mammalian organism on a commercially viable scale.

    We utilize proprietary molecular biology systems to more efficiently knockout genes in mice on a large scale and conduct a detailed analysis of the resulting physiological, pathological and behavioral effects in these mice. As a result, we assess the function of the gene in a mammal that is closely related genetically and physiologically to humans.

6

    We believe our technology platform and approach offers significant advantages over the standard and genetic approaches, including:

•

RAPIDLY GENERATING INTELLECTUAL PROPERTY ON THE IN VIVO MAMMALIAN FUNCTIONAL ROLE OF GENES AND SECRETED PROTEINS. We are pursuing intellectual property protection for our gene function discoveries and under certain of our programs, plan to grant our customers the right to use our intellectual property. Although we have filed almost 200 applications with the United States Patent and Trademark Office, we do not currently hold any issued patents.



•

INCREASING THE SCALE AND SPEED OF GENERATING MAMMALIAN GENE FUNCTION INFORMATION, DELIVERING VALIDATED GENE TARGETS AND DISCOVERING POTENTIAL SECRETED PROTEIN DRUG CANDIDATES. Based on our current knockout rates, we expect to be able to target, analyze and deliver detailed in vivo gene function information on approximately 250 to 300 different genes per year. We additionally expect to target and analyze approximately 100 to 150 secreted protein gene knockouts per year. This is a significant improvement over traditional approaches that produced an estimate of over 700 total worldwide reported mouse knockouts between 1991 and 1997. Our proprietary high-throughput gene knockout and analysis system can be scaled-up to greater capacity if we determine additional production is required.



•

REDUCING THE COST OF DETERMINING GENE FUNCTION, PROVIDING VALIDATED GENE TARGETS AND IDENTIFYING POTENTIAL SECRETED PROTEIN DRUG CANDIDATES. By providing a fully integrated target validation system as opposed to a multi-tier process, we believe we can reduce the number of steps and costs associated with the target validation process and the number of parties to whom royalties must be paid. Through our DeltaBase™ database product, we provide our subscribers with information on gene function and the target potential of genes earlier in the drug discovery process than under the standard or genetic approaches. We believe that early access to IN VIVO data allows selection of appropriate drug targets, increases efficiency and reduces costs by allowing our subscribers to focus on genes with high potential for successful drug development. This information may allow our subscribers to eliminate non-viable targets from potential development earlier in the discovery process. Our fully integrated target validation system may also increase the efficiency and reduce the costs associated with our discovery of potential secreted protein drug candidates in our collaborative and our own internal secreted protein programs.



•

PRE-SELECTING COMMERCIALLY RELEVANT MAMMALIAN GENE TARGETS. We have focused our target validation efforts on gene families that we believe have the greatest potential for drug development. Worldwide genome sequencing efforts have identified many new members of the gene families currently targeted by the pharmaceutical and biotechnology industry, including approximately 1,700 members that we have initially selected that may have relevance to disease and are potential targets of drug discovery efforts.



•

PROVIDING ACCESS TO KNOCKOUT MAMMALIAN ANIMAL MODELS. The preclinical testing, or animal testing, of drugs has often been impeded by the lack of animal models that can represent the human disease condition. We believe our fully integrated target validation system can produce and deliver relevant knockout mouse models to the pharmaceutical and biotechnology industries. These knockout mouse models can be used for further research and development relating to gene function and disease analysis.



•

ALLOWING OUR CUSTOMERS TO STORE, ACCESS, MANIPULATE AND ANALYZE GENE FUNCTION INFORMATION THAT WE GENERATE. We have developed a proprietary information technology infrastructure for the delivery, maintenance and use of the data we produce. Each year under our DeltaBase program alone, we expect to generate over

7

five million individual measurements of physiological, biochemical and behavioral observations, or data points, relating to gene function in mammalian systems. We organize and will deliver our data in a manner that we believe will provide simple and rapid accessibility. Additionally, our data is compatible with standard computing tools used by the pharmaceutical and biotechnology industries.

    In addition to our DeltaBase™ gene function database program, we have our own internal and collaborative programs to discover novel, commercially relevant secreted proteins. Secreted proteins are proteins that play an important role in the formation, regulation, growth and maintenance of multi- cellular organisms. Examples of well-known secreted proteins discovered by other companies include insulin, human growth hormone and erythropoeitin, or EPO. Using our powerful technology platform along with our other proprietary technologies, we have developed a program that identifies and defines the mammalian IN VIVO function of mammalian secreted proteins. Specifically, we have proven genetic technologies that allow us to more rapidly identify and knockout secreted proteins in mice. We believe that our secreted protein discovery programs provide a foundation for developing and commercializing proprietary therapeutic protein products.

Our Strategy

    Our goal is to be the leading provider of data on the function, role and disease relevance of mammalian genes to the pharmaceutical and biotechnology industries. We believe our data will improve the speed, efficiency and effectiveness of drug discovery. Additionally, we intend to use our powerful technology platform along with our other proprietary technologies to discover novel therapeutic secreted proteins to provide a pipeline of potentially new biotechnology drug candidates for collaborative programs or our own internal development, or in alliance with strategic partners, we may also choose to in-license potential secreted protein targets. As we have a limited operating history and an unproven business strategy, we cannot assure you that we will succeed in achieving our goal. The key elements of our strategy include:

•

BECOMING THE MOST COMPREHENSIVE SOURCE OF MAMMALIAN INFORMATION ON GENE FUNCTION AND TARGET VALIDATION. We intend to expand our current technology platform and develop new programs and systems to increase the scale, scope and depth of our ability to determine mammalian gene function. We believe that we can continue to discover and deliver information that is relevant to the drug discovery process.



•

PURSUING THE DISCOVERY AND EARLY-STAGE DEVELOPMENT OF POTENTIAL SECRETED PROTEIN DRUG CANDIDATES FOUND THROUGH OUR SECRETED PROTEIN PROGRAM. We plan to continue the development of our secreted protein discovery program in order to provide a pipeline of secreted proteins to serve as potential drug discovery candidates. Currently, we plan to target, knockout and analyze approximately 500 secreted protein genes in knockout animal models to discover, identify and assess the function of these secreted proteins. We intend to pursue development of selected opportunities that arise from this program and in-licensing opportunities. This development may be internal or in collaboration with other pharmaceutical and biotechnology companies.



•

FOCUSING ON THE COMMERCIAL NEEDS OF OUR CUSTOMERS. We plan to deliver valuable gene function information to our customers and allow them to concentrate on the drug discovery process downstream of target validation. By focusing our research process on target validation and obtaining functional information, we believe we provide our customers meaningful time and cost savings in their drug discovery efforts.



•

CONTINUING TO PURSUE INTELLECTUAL PROPERTY RIGHTS. We are employing an intellectual property strategy to secure patent, trademark and copyright protection for what we believe to be our commercially relevant inventions, products, and methods. Under certain

8

programs and collaborations, we intend to offer our customers access to certain of our intellectual property rights. Although we have filed over 200 applications with the United States Patent and Trademark Office, we do not hold any issued patents. We believe that if we are able to secure patent rights around gene function and functional utility associated with the commercialization of genes, secreted proteins and our animal models as drug targets, we may be able to generate licensing and other revenues associated with such rights.

•

CONTINUING OUR EARLY-STAGE BIOPHARMACEUTICAL PRODUCT DEVELOPMENT PROGRAMS. We recently in-licensed from the University of Kentucky, rights to research, develop and commercialize methods and compounds targeting CD123, a unique marker for certain types of leukemic stem cells, as a potential treatment of Acute Myelogenous Leukemia (AML). We plan to continue our early-stage evaluation and development of this potential approach for the treatment of AML and expect to continue advancing this program through collaborative or other efforts.



•

ACQUIRING TECHNOLOGIES TO MEET THE CONTINUING TARGET VALIDATION NEEDS OF CUSTOMERS. We intend to acquire and license additional products and programs, if we determine that these products or programs complement our existing target validation technologies or augment our existing information technology platforms.

Our Products and Programs

    We have developed and plan to continue to develop technologies, products and programs that determine the function and disease relevance of genes in mammalian organisms. We have developed and are expanding the following products and programs:

DeltaBase™

Overview

    DeltaBase is our database that provides information, based on knockout mouse studies, on gene function and validated gene targets for drug discovery. We created DeltaBase to be marketed to the pharmaceutical and biotechnology industries to help define the role that genes play in biological processes and disease. We believe that DeltaBase is a valuable resource for mammalian gene function information and validated targets.

    At our current rate of production, we expect to provide gene function and target validation information through DeltaBase, on approximately 250 different mammalian genes per year. We select genes for DeltaBase based upon what we believe to be their potential to become useful drug targets. We generate information on these genes by comprehensively analyzing knockout mice generated through our proprietary, gene knockout methods. Each knockout mouse undergoes a standardized, detailed and extensive analysis in order to determine the function and role that a particular gene plays in the mouse and that gene's suitability as a drug target. We currently intend that DeltaBase will deliver a total of approximately five million individual, IN VIVO function, data points per year. We believe that the body of gene function information delivered under DeltaBase provides an advantage to the drug discovery efforts of pharmaceutical and biotechnology companies by reducing the time required for target validation.

    In addition to accessing target validation data, DeltaBase subscribers will have access to the knockout mice used to generate this data. Access to these animals will allow DeltaBase subscribers to more rapidly pursue specific areas of interest. We believe this will be attractive to smaller pharmaceutical companies who lack the necessary infrastructure for wide-scale target validation programs.

9

DeltaBase Technologies

    We designed DeltaBase to provide our subscribers with the ability to compare resulting phenotypic and gene function data across hundreds of different mammalian genes from different gene families selected for their potential commercial relevance to drug discovery. In order to generate, analyze, store, manipulate and deliver such large volumes of data and information, we have developed proprietary, high-volume, assembly-line methods to:

•

RAPIDLY AND EFFICIENTLY GENERATE LARGE NUMBERS OF KNOCKOUT MICE ANIMAL MODELS. We utilize proprietary molecular biology systems to more efficiently knockout genes in mice on a large scale. We are able to move directly from a small amount of gene sequence information straight to the production of knockout mice and the determination of gene function and validated targets. Each stage of our knockout generation process has been scaled-up for high-throughput production to generate and analyze approximately 250 gene knockout animal models per year for our Deltabase program, which we believe is a significant improvement over historical, relatively limited industry production. This system can be readily scaled up to even greater capacity if we determine additional production is required.



•

SELECT MAMMALIAN GENES AND GENE FAMILIES TO IDENTIFY VALIDATED DRUG TARGETS. In selecting the gene families for DeltaBase, we targeted those that have demonstrated their value as drug development targets, have led to the commercialization of successfully marketed drugs and present potential additional drug development targets. The current gene families represented in DeltaBase include the G-protein coupled receptors, ion channels and proteases. As part of the gene target selection process, we utilize information technology, statistical analysis and biological information systems to extract and analyze publicly available data on the human genome to search for additional genes and gene families with potential commercial relevance to drug discovery efforts. This application of statistical and mathematical models to genetics is known as bioinformatics. To date, our bioinformatics program has focused on an initial pipeline of approximately 1,700 potential targets for development under DeltaBase that we believe may be of interest to prospective pharmaceutical and biotechnology subscribers.



•

EXTENSIVELY ANALYZE THE KNOCKOUT MICE GENERATED. We have developed and employ large-scale assembly-line analysis programs that provide detailed physiological, pathological and behavioral data. This analysis is performed on all tissues and organ systems within the mouse. Moreover, this analysis of the entire organism may provide information on possible side effects and toxicology profiles associated with each gene and its function. We are in the process of developing detailed analytical programs in areas such as the central nervous system, cardiovascular system, infectious disease, inflammation and the immune system. We believe that the DeltaBase knockout mice can serve as efficient vehicles for the generation of additional complementary information and data on gene function.



•

ACCURATELY AND EFFICIENTLY CAPTURE, STORE, MANIPULATE AND DELIVER APPROXIMATELY FIVE MILLION INDIVIDUAL DATA POINTS PER YEAR GENERATED FROM THE ANALYSES OF KNOCKOUT MICE. DeltaBase subscribers will have the ability to access, utilize and perform multifaceted analysis on the gene function data and information contained in DeltaBase. In addition, our proprietary information technology allows our customers to perform searches of the gene function analyses contained in DeltaBase, obtain detailed scientific and pathology summaries of gene function findings and submit inquiries and questions to DeltaBase through a medical/scientific vocabulary search engine containing approximately over one million terms. To meet the needs of DeltaBase subscribers, the data and information is readily exportable and can be manipulated by information technology tools and other databases widely employed in the pharmaceutical and biotechnology industries.

10

Marketing and Customer Agreements

    The Company has DeltaBase agreements with both GlaxoSmithKline plc and Pfizer Inc. Each of these agreements provides for payments aggregating $15 million over three years. Under the DeltaBase agreements, GlaxoSmithKline and Pfizer have the right to access DeltaBase information on gene function and validated gene targets based upon knockout mouse studies. The DeltaBase agreements also grant certain non-exclusive, worldwide licenses to knockout mice, materials and intellectual property rights under DeltaBase. In addition, we may receive additional fees for access to our intellectual property rights. We do not expect the additional licensing fees payable under these agreements to be material. We could also receive additional payments based upon the achievement of designated milestones. We cannot assure you that we will receive any milestone payments since payments are entirely dependent upon the research, development and commercialization of products by GlaxoSmithKline and Pfizer. GlaxoSmithKline may terminate our agreement for any reason within the first three months after the one-year anniversary of the effective date upon payment of a specified termination fee.

DeltaSelect™

Overview

    DeltaSelect is our custom gene knockout program that uses our powerful platform technology employed in our DeltaBase database program. Our DeltaSelect program is different, however, because our customers select and identify to us the particular genes that they wish to have knocked out in mice. We provide customers with access to our gene knockout technologies and the resulting knockout mice, data and information generated under each DeltaSelect program. We believe that this program has provided validation of our proprietary platform technology and promoted interest in DeltaBase. We anticipate that revenues from DeltaSelect will continue to become less significant and that DeltaSelect will be utilized under limited circumstances to develop new technologies, product offerings and programs in collaboration with pharmaceutical companies.

    We have produced customized knockout mice at the direction of our customers for a limited number of pharmaceutical companies. Currently, we are undertaking DeltaSelect knockout programs under agreements with GlaxoSmithKline, Merck & Co., Inc. and Schering-Plough Research Institute.

DeltaSelect Technology

    In addition to the proprietary platform technology developed by us, we are currently employing and developing employing additional technologies that can be used to create conditional knockout mice. Conditional knockout mice are mice where the gene of interest is removed under unique conditions in a specific tissue or cell type at selected and controlled times. We are currently developing conditional knockout systems for our DeltaSelect program using CRE/LOX and FLP/FRT recombinase technologies.

Secreted Protein and Biopharmaceutical Program

Overview

    Secreted proteins represent proteins that are synthesized for export from the cell or to the surface membrane of the cell where they play a role in the communication between cells. These communication roles are essential for the formation, regulation, growth and maintenance of multi-cellular organisms. Currently, secreted proteins constitute the majority of successful targets for drug discovery. Examples of well-known secreted proteins discovered by other companies include insulin, human growth hormone and erythropoeitin, or EPO.

11

    The goal of our secreted protein program is to provide a potential pipeline of validated new biotechnology drug candidates either for internal development or development in alliance with strategic partners. We first identify new therapeutic proteins that have the potential to become drugs. This is through our targeted genomics and proprietary bioinformatics technologies, our proprietary gene trap technologies licensed from the University of Edinburgh or in alliance with strategic partners. Once a target is identified, we determine the function of the secreted protein through our proprietary knockout platform technologies and phenotypic analysis programs. The resulting secreted protein discoveries have the potential to become candidates for further drug development. In addition, we also plan to acquire rights to additional potential secreted protein drug candidates through collaborations or alliances with others.

Gene Trap Technology

    We are an exclusive worldwide licensee of a secreted protein gene trap from the University of Edinburgh that identifies genes that code for secreted proteins and simultaneously enables the production of knockout mice to determine the in vivo function of these genes. This secreted protein gene trap technology will work in all cell types. However, when the system is employed in mouse ES cells, the resultant ES cells containing the gene deletion can be used to more rapidly generate knockout mice.

    We believe that the benefits of this proprietary technology are its ability to:

•

identify rarely turned on, or expressed, genes;



•

move directly to IN VIVO analysis in the mouse eliminating the need for cell-based experiments that require protein or antibody production to define function;



•

directly define the role of cell surface receptors and function in all cell types; and



•

identify key secreted proteins in a variety of physiological conditions.

CD 123 Antigen

    In November 2000, we entered into an exclusive worldwide license agreement with the University of Kentucky to research, develop and commercialize methods and compounds targeting CD123, a unique marker for certain types of leukemic stem cells. CD123 was recently discovered to be a unique marker for human acute myelogenous leukemia stem cells. Recent studies suggest that the population of malignant cells found in AML arises from a rare population of these leukemic stem cells that express the CD123 protein at high levels. We believe that using CD123 as the target, it may be possible to design a new drug or antibody that will selectively kill only leukemic stem cells. We plan to continue our early-stage evaluation and development of this potential approach for the treatment of AML and expect to continue advancing this program through collaborative or other efforts. It is our current plan to initiate pre-clinical development of a CD123 antibody for the treatment of AML.

DeltaXpress™

    We currently have under development a new database product for gene expression analysis. If we are successful in developing and making this product commercially available, we anticipate that we would offer customers subscription agreements for the non-exclusive use of this database.

    We have entered into an agreement with Affymetrix, Inc. for the supply of products for gene expression analysis. Our agreement with Affymetrix is only the first of many steps necessary to create this potential additional product.

    These products allow for determination of the level of activity of particular genes in a particular tissue, or gene expression. The activity of a gene or a set of genes can differ based upon the

12

physiological state within an organism. For example, the presence or absence of certain genes from a set of genes that interact with each other can cause the activity level of the other genes in that set to increase or decrease. This gives clues as to the functional roles and interactions of these genes.

    Analysis of gene knockout mice in our gene function database provides detailed functional information on the role of individual genes. However, this information can be expanded to encompass the potential functional role of many additional genes by using expression analysis of tissues from the knockout mice. In knockout mice, the deletion of a gene creates a situation wherein the animal attempts to compensate for the loss of the particular gene and therefore activates or suppresses the expression of functionally related genes. We believe that these systematic gene expression analysis studies of our knockout mice may provide a method for assembling functional maps for large segments of the genome. Therefore, we believe that DeltaXpress may add significant value to the genomic content of our databases.

Customers

    In 2000, we entered into DeltaBase Collaboration Agreements with GlaxoSmithKline plc and Pfizer Inc. that provide these companies the right to access DeltaBase information on gene function and validated gene targets.

    Under our DeltaSelect program, we have entered into arrangements with major pharmaceutical companies where we produce customized standard, or unconditional, knockout mice. We currently perform services under our DeltaSelect program for Schering Plough Research Institute, Merck & Co., Inc. and GlaxoSmithKline plc.

    As we are in the early stages of development of our DeltaXpress program, we do not have any customers for this program.

    GlaxoSmithKline and Schering Plough accounted for 56% and 18%, respectively, of our revenues in 2000.

Research and Development

    As of December 31, 2000, we had a total of 179 employees dedicated to research and development activities. We have spent substantial funds over the past three years to develop our database and other programs and expect to continue to do so in the future. Research and development expenses were $26.3 million in 2000, $12.1 million in 1999 and $3.4 million in 1998.

Intellectual Property

    Our policy is to pursue patent protection around our commercially relevant products, techniques and methods. Although we do not currently hold any issued patents, we intend to file applications covering all the knockout mice we produce. We also intend to pursue patent, copyright and trademark protection with respect to any information technologies, systems or other products that we believe would benefit from these protections. We cannot assure you, however, that any of our applications on file with the United States Patent and Trademark Office will result in the issuance of any patents, that our patent applications will have priority over others' applications, or that, if issued, any of our patents will offer protection against our competitors. Additionally, we cannot assure you that any patent issued to us will not be challenged, invalidated or circumvented in the future or that the rights created thereunder will provide a competitive advantage. Litigation may be necessary to enforce any patents issued to us, to protect trade secrets or know-how owned by us or to determine the enforceability, scope, and validity of the proprietary rights of others.

    Others may have filed and in the future are likely to file patent applications that are similar or identical to ours. To determine the priority of inventions, we may have to participate in interference

13

proceedings declared by the U.S. Patent and Trademark Office that could result in substantial cost to us. We cannot assure you that any patent application of another will not have priority over patent applications filed by us. Our commercial success depends in part on our neither infringing patents or proprietary rights of third parties nor breaching any licenses that may relate to our technologies and products.

    We have obtained licenses for certain technologies. However, we cannot assure you that we will be able to obtain licenses for technology patented by others on commercially reasonable terms, if at all, that we will be able to develop alternative approaches if unable to obtain licenses or that our current and future licenses will be adequate for the operation of our business. Our failure to obtain necessary licenses or to identify and implement alternative approaches could have a material adverse effect on our business, financial condition and results of operations.

    We also rely upon trade secrets, technical know-how and continuing invention to develop and maintain our competitive position. We cannot assure you that others will not independently develop substantially equivalent proprietary information and techniques or otherwise gain access to our trade secrets or disclose such technology, or that we can meaningfully protect our trade secrets, or that we will be capable of protecting our rights to our trade secrets.

    On May 24, 2000, Lexicon Genetics Incorporated filed a lawsuit against us in the United States District Court for the District of Delaware. The complaint in the lawsuit alleges that our methods of making knockout mice infringe United States Patent No. 5,789,215, under which Lexicon claims to be an exclusive licensee. The complaint seeks a judgment that we infringed this patent, a permanent injunction against further infringement of the patent and an award of damages in an unspecified amount that, under certain circumstances, may be trebled. On June 13, 2000, we responded to Lexicon's complaint by filing an answer and seeking a declaratory judgment in our favor. On October 31, 2000, an Order was entered by the Court amending the complaint and adding GenPharm International, the patentee of the '215 patent, as a plaintiff. On November 14, 2000, we filed an answer as well as federal antitrust counterclaims under the Sherman Antitrust Act against Lexicon and GenPharm, alleging that Lexicon asserted a fraudulently procured patent (the '215 patent) against us in an attempt to illegally monopolize the market for production and sale of knockout mice, a violation of Section 2 of the Sherman Act. The counterclaim asserts that Lexicon knew that the '215 patent was obtained by fraud and that the patent was invalid when it brought suit against us. The antitrust counterclaim also charges Lexicon and GenPharm with conspiracy to monopolize the market for knockout mice, in violation of Section 2 of the Sherman Act, as well as with concerted conduct in restraint of trade, in violation of Section 1 of the Sherman Act. The counterclaim seeks damages including lost profits, devaluation of the initial public offering and legal fees, as well as a trebling of damages pursuant to Section 4 of the Clayton Act. In addition, the answer seeks a declaratory judgment that we do not infringe the claims of the '215 patent and that the patent is invalid. The answer also asserts that the patent was obtained by knowing and willful fraud on the U.S. Patent and Trademark Office rendering the patent unenforceable. We intend to defend the action vigorously.

    The litigation is in the early stages and its outcome cannot be predicted. We believe that if Lexicon were to prevail in this lawsuit, the impact would not significantly affect our operations. We believe that the methods covered by the '215 patent represent a modification of prior methods of generating knockout mice. Although we believe this modification was previously described by others prior to GenPharm's filing of its patent application, we have intentionally avoided the use of this modification.

    In addition, on October 13, 2000, Lexicon and the University of Utah Research Foundation filed a lawsuit against us in the United States District Court for the Northern District of California. The complaint in this lawsuit alleges that we are infringing United States Patents Nos. 5,631,153, 5,464,764, 5,627,059 and 5,487,992 (the Capecchi patents), under which Lexicon claims to be an exclusive licensee. All of these patents relate to the use of a positive-negative selection vector in gene targeting. The

14

complaint seeks a judgment that we infringed these patents, a preliminary and permanent injunction against further infringement of the patents and an award of damages in an unspecified amount that, under certain circumstances, may be trebled. On November 8, 2000, we filed an answer denying infringement, and a counterclaim seeking a declaratory judgment that the patents are invalid. The counterclaim seeks an award of damages including attorney fees.

    We believe that the methods employed by us do not infringe these patents and that the patents are invalid. We believe that the Capecchi patents cover certain modifications to existing methods of generating knockout mice, such modifications that we do not practice. These modifications are therefore but one method available for use in generating knockout mice. Numerous equally effective design-arounds exist, as well as methods that predate the Capecchi patents that are in the public domain and thus outside the scope of the patent. Thus, if Lexicon prevails in any such lawsuit, we would only be prevented from using certain of its methods but would not be prevented from using all of our current methods. However, an adverse determination as to those certain methods could significantly affect the development and marketing of our existing function database and DeltaSelect programs and could cause us to incur significant financial liabilities that could materially affect our business and operating results.

    At this time no estimate can be made of the outcome or any possible loss that may be incurred. The litigation is in the early stages and its outcome cannot be predicted.

Competition

    We face significant competition in the area of genomics-based research from for-profit companies such as Celera Genomics, Curagen, Inc., DNX (a subsidiary of Phoenix International Life Sciences, Inc.), GeneLogic, Inc., Human Genome Sciences, Inc., Incyte Pharmaceuticals, Inc., Lexicon Genetics Incorporated and Millennium Pharmaceuticals, Inc., among others, many of which have substantially greater financial, scientific and human resources than we do. In addition, the Human Genome Project and a large number of universities and other not-for-profit institutions, many of which are funded by the U.S. and foreign governments, are also conducting research to discover genes and their function.

    We face, and will continue to face, significant competition in our efforts to validate drug targets and to attract research dollars. Many other companies have or are developing capabilities in the use of living organisms to define gene function. These competitors include such companies as Lexicon Genetics Incorporated, Exelixis, Inc., and Devgen N.V. Additionally, many genomics companies may expand their capabilities to determine gene function. We also believe that some pharmaceutical and biotechnology companies are discussing the possibility of working together to discover the functions of genes and share gene function-related data among themselves. The formation of this type of consortium could reduce the customer base for our gene function-related business. Further, as we expand our range of products and services, such as our secreted protein program, we will compete with additional companies, some of which may be our customers at that time or potential customers.

    Companies focused specifically on other organisms, such as fruit flies, worms and yeast, use methods of identifying potential drug targets which are different than ours. In addition, pharmaceutical, biotechnology and other genomics companies, as well as a number of universities and other not-for-profit institutions, are seeking to develop competing technologies. Many of these competitors have substantially greater financial, scientific and human resources than we do. Many of these competitors also have substantially greater experience than we do in their respective fields. As a result, our competitors may succeed in developing products and technologies earlier than we do or in developing products and technologies that are more effective than ours.

    We believe that the principal competitive factors in selling our products and services are the quality and reliability of the gene function information, the volume of the gene function information,

15

the features and ease of use of database products and the cost and pricing of competing products. We believe that we compete favorably with respect to these factors; however, our market is rapidly changing and we expect to face further competition from new market entrants and consolidation of our existing competitors.

Government Regulation

Regulation of Animal Use

    The federal Animal Welfare Act, or AWA, governs the humane handling, care, treatment and transportation of some animals used in U.S. research activities. Rats, birds and mice, including the mice in our knockout programs, are currently not subject to regulation under the AWA. However, the United States Department of Agriculture, which enforces the AWA, has been sued on this matter and recently agreed, as part of the settlement of this lawsuit, to begin the process of changing the regulations issued under the AWA to include rats, mice and birds within its coverage. Congress subsequently prohibited, for the next fiscal year ending September 30, 2001, the expenditure of any money for the purpose of changing the regulations with respect to including rats, mice and birds. The AWA imposes a wide variety of specific regulations on producers and users of animal subjects, most notably personnel, facilities and statistical standards, cage size, feeding, watering and shipping conditions and environmental enrichment methods. If the USDA decides to include mice in its regulations, we could be required to alter our production operation for these models, including adding production capacity, new equipment and additional employees. It is possible that the USDA's actions will negatively affect our operations. In addition, although we do not anticipate the addition of mice to the AWA to require significant expenditures, it is possible that the AWA, when amended, may be more stringent than we expect and require significant expenditures. Any future amendments to the AWA or other laws or regulations may also require significant expenditures by us.

    Furthermore, some states have their own regulations, including general anti-cruelty legislation, which establish certain standards in handling animals. To the extent that we provide products and services overseas, we also have to comply with foreign laws, such as the European Convention for the Protection of Animals During International Transport and other anti-cruelty laws. The Council of Europe is presently considering proposals to more stringently regulate animal research.

Regulation of Genetically Modified Organisms

    Since we are in the business of developing animals containing changes in their genetic make-up, we may become subject to a variety of laws, guidelines, regulations and treaties specifically directed at genetically modified organisms, or GMOs. The area of environmental releases of GMOs is rapidly evolving and is currently subject to intense regulatory scrutiny, particularly internationally. Current laws, guidelines and other requirements typically include confinement requirements for preventing the spread of GMOs into the environment. Examples of these guidelines in the U.S. include the National Institutes of Health "Guidelines for Research Involving Recombinant DNA Molecules" and the USDA "Guidelines for Research Involving the Planned Introduction into the Environment of Genetically Modified Organisms". Although these guidelines typically apply only to federally-funded activities, if we were to become subject to similar laws in the future, we could incur compliance costs.

    The Biosafety Protocol, or the BSP, is also of particular importance to our international operations. The BSP, a treaty recently adopted in Montreal, Canada in late 1999, is expected to be ratified in many countries internationally in approximately two years. Many industrialized and non-industrialized countries will be signatories to the BSP. Although the U.S. is not subject to the BSP, if ratified, the BSP is expected to cover shipments from the U.S. to countries abroad that have signed the BSP. The BSP is also expected to cover the importation of living modified organisms, a category that could include our

16

animals. If our animals are not contained as described in the BSP, our animals could be subject to the potentially extensive import requirements of countries that are signatories to the BSP.

Regulation of New Drugs or Biological Drugs for Human Use

    We are planning to engage in the development and commercialization of therapeutic products that will be regulated by governmental authorities in the United States and other countries. Those governmental authorities, including the Food and Drug Administration (FDA), extensively regulate the testing, manufacturing, labeling, advertising, promotion, export and marketing, among other things, of therapeutic products. Any new therapeutic product administered to human patients is regulated as a drug or a biological drug and requires regulatory approval before it may be commercialized.

    In the United States, our potential therapeutic products likely will be regulated as human drugs, including biological drugs. The FDA will require us to file and obtain approval of a Biologics License Application (BLA) (for biologics) or a New Drug Application (NDA) (for other drugs) before we can commercialize any such products. BLAs and NDAs cover both the facility in which the products are manufactured and the products themselves. Generally, biological drug regulation is more rigorous than other drug regulation, particularly because biologics are subject to lot-to-lot release requirements whereas other drugs are not.

    The testing and approval process requires substantial time, effort and financial resources. After approval is obtained, a supplemental approval is generally required for each proposed new indication and for many different types of manufacturing changes. The supplement often contains data similar to that submitted in the original BLA or NDA.

    Any future product approvals that are granted remain subject to continual FDA review, and newly discovered or developed safety or efficacy data may result in withdrawal of products from marketing. Moreover, if and when such approval is obtained, the manufacture and marketing of future products will remain subject to extensive regulatory requirements administered by the FDA and other regulatory bodies, including compliance with current Good Manufacturing Practices, adverse event reporting requirements and the FDA's general prohibitions against promoting products for unapproved or "off-label" uses. Companies are subject to inspection and market surveillance by the FDA for compliance with these regulatory requirements. Failure to comply with the requirements can, among other things, result in warning letters, product seizures, recalls, fines, injunctions, suspensions or withdrawals of regulatory approvals, operating restrictions and criminal prosecutions.

    Companies also are subject to a variety of regulations governing clinical trials and sales of their products outside the United States.

Other Regulation

    We are also subject to a variety of other federal and state laws and regulations in the U.S. and in other countries pertaining to our facilities, the shipment, exportation and importation of various articles and health and safety matters.

Employees

    As of December 31, 2000, we had a total of 212 full-time employees. Of these, 36 hold Ph.D.s and 10 hold other advanced degrees. None of our employees is represented by a labor union. We consider our relations with our employees to be good.

Executive Officers of Registrant

    William Matthews, Ph.D., a co-founder of our company, has served as our President since February 1997 and as our Chief Executive Officer since December 1998. Dr. Matthews has served as a

17

director of Deltagen since February 1997. Prior to founding our company, Dr. Matthews worked at Genentech, Inc., from June 1992 to January 1997 where he established and ran a program in stem cell biology. Dr. Matthews received his Ph.D. in cell biology from Southwestern Medical School in Dallas, followed by post-doctoral fellowships at Harvard Medical School and Princeton University.

    Mark W. Moore, Ph.D., a co-founder of our company, has served as our Chief Scientific Officer and Treasurer since February 1997. Prior to founding our company, Dr. Moore worked from August 1991 to January 1997 at Genentech, Inc. where he established and directed Genentech's gene knockout program in mice. Dr. Moore was a Leukemia Society of America post-doctoral fellow in molecular and cellular immunology in the laboratory of Dr. Michael Bevan at Scripps Clinic. Following his post-doctoral work, Dr. Moore served on the faculty of the Norris Cancer Center at the University of Southern California. Dr. Moore received his A.B. in biochemistry from Princeton University and his Ph.D. in biology from Brandeis University.

    Augustine G. Yee has served as our Vice President of Corporate Development, General Counsel and Secretary since April 1999. Prior to joining us, Mr. Yee was a patent and intellectual property litigation attorney with the law firm of Lyon & Lyon LLP from October 1993 to July 1995, and a corporate securities and technologies attorney with Pillsbury Madison & Sutro LLP from August 1995 to April 1999, where he represented companies in the biotechnology and information technology fields. Mr. Yee was formerly a law clerk to the Honorable Edward Rafeedie, United States District Court, Central District of California, and is admitted to practice before the United States Patent and Trademark Office. Mr. Yee received his B.S. in molecular biology from the University of California at San Diego and his J.D. from Pepperdine University School of Law.

    Lars Barfod served as our Vice President of Commercial Development from October 1999 until March 2001. From March 1999 to September 1999, Mr. Barfod served as Vice President, Sales & Marketing at Ciphergen Biosystems, Inc. From April 1997 to January 1999, Mr. Barfod served as Vice President of Marketing at Genentech. Mr. Barfod was also a member of Genentech's Product Development and commercial Operations committees. Prior to joining Genentech, Mr. Barfod spent 11 years at Novo Nordisk Pharmaceuticals, Inc., most recently as Vice President of Sales, Marketing and Business Development. Mr. Barfod received his Masters degree in marketing from the Royal University of Copenhagen, Denmark.

    John E. Burke has served as our Vice President of Intellectual Property since December 1999. Prior to joining us, Mr. Burke was Of Counsel with the law firm of Pillsbury Madison & Sutro LLP from 1996 to 1999. Prior to that time, he was a patent and intellectual property attorney with the law firm of Schwegman, Lundberg, Woessner & Kluth from 1995 to 1996, and served as Corporate Patent Counsel for Cortech, Inc., from 1993 to 1995. Mr. Burke was also a patent attorney with the law firm of Morgan & Finnegan from 1990 to 1992. Mr. Burke is admitted to practice before the U.S. Court of Appeals for the Federal Circuit, the U.S. Supreme Court, and the U.S. patent and Trademark Office and is a member of the California and New York state bars. Mr. Burke received his B.S. in chemical/biochemical engineering from Rutgers College of Engineering and his J.D. from Rutgers School of Law.

    Terry Coley, Ph.D., has served as our vice President of Information Technology since September 1999. Prior to joining us, Dr. Coley was co-founder and CEO of Virtual Chemistry, Inc., from January 1996 to August 1999. At Virtual Chemistry, Dr. Coley established software teams to engineer custom software for biotechnology and pharmaceutical companies. Prior to that time, Dr. Coley worked as a molecular modeling software development project leader at Molecular Simulations Inc. Dr. Coley received his B.S. in chemistry and computer science from the University of Illinois and his Ph.D. in computational chemistry from the California Institute of Technology.

    Robert Klein, Ph.D., has served as Vice President of Technology Development since June 2000. Prior to that time, Dr. Klein held positions as Director of Molecular Biology from June 1998 through

18

June 2000 and Senior Scientist from May 1997 through June 1998 at Deltagen. Prior to joining Deltagen, Dr. Klein worked at Genentech, Inc. from August 1993 through May 1997 where he was involved in Genentech's functional genomics program. Dr. Klein also served as a project team leader for Genentech's lead protein therapeutic for treatment of Parkinson's disease. Dr. Klein received his A.B. in biochemistry from the University of California at Berkeley and his Ph.D. in biology from the Massachusetts Institute of Technology.

    Richard H. Hawkins has served as our Chief Financial Officer since September 2000. Prior to joining us, Mr. Hawkins was an independent consultant. From March 1984 until July 1999, Mr. Hawkins was employed by McKesson Corporation (McKesson HBOC, Inc.) where he served as Chief Financial Officer from September 1996 until July 1999. Mr. Hawkins received his B.S. in Chemistry from Stanford University and his M.B.A. from the University of Chicago. He is a licensed Certified Public Accountant.

    Brian E. Crowley has served as our Director of Finance since August 1999. Prior to joining us, Mr. Crowley was the Director of Finance for the Dataconferencing product division at Polycom, Inc. from January 1997 to February 1998 and the General Accounting Manager from November 1994 to January 1997. Prior to that time, Mr. Crowley served as controller at The LAN Guys Inc. and Conductus, Inc., and as an associate at PricewaterhouseCoopers LLP. Mr. Crowley received his B.S. from St. Mary's College of California and his M.B.A. from the University of Notre Dame. He is a licensed Certified Public Accountant.

    Paul Laland has served as our Vice President of Corporate Communications since February 2001. Prior to joining us, Mr. Laland was executive vice president and head of Health Technology at GCI Group, a public and investor relations practice dedicated to biotechnology and genomics, from 1999 to February 2001. From 1995 to 1999 Mr. Laland served as director of Corporate Communications at Genentech, Inc. and from 1992 to 1995 was associate director of Corporate Communications at Synergen, Inc. Mr. Laland holds a B.A. in Communications from University of Utah.

    Kay Slocum has served as our Vice President of Human Resources since March 2001. Prior to joining us, Ms. Slocum was Vice President of Human Resources at Coulter Pharmaceuticals, Inc. from 1996 until February 2001. She served as an independent consultant from 1995 to 1996. From 1993 to 1995 Ms. Slocum was Manager, Corporate Employee Development of Varian Associates, Inc. Ms. Slocum holds a B.A. in Sociology from Southern Illinois University and a M.S. in Industrial Relations from Loyola University of Chicago.

    Dan Shochat, Ph.D. has served as our Vice President of Pharmaceutical Development since February 2001. Prior to joining us, Dr. Shochat was employed by Coulter Pharmaceutical where he was Senior Vice President and Chief Financial Officer since 1998 and Vice President, Research and Development from March 1995. From July 1988 to April 1995, Dr. Shochat served as Director of Biotechnology Development at the Medical Research division of American Cyanimid, Inc., where he was responsible for the worldwide program in monoclonal antibodies for the treatment of cancer. He received his B.S. and M.S. degrees from Hebrew University in Israel and a Ph.D. in Biochemistry from L.S.U. Medical School in New Orleans.

19

RISKS

We expect to continue to incur substantial losses and we may never achieve profitability, which in turn may harm our future operating performance and may cause the market price of our stock to decline.

    We have had net losses every year since our inception in 1997 and, as of December 31, 2000, had an accumulated deficit of $50.7 million. We had net losses of $32.2 million, $13.8 million and $3.4 million in 2000, 1999 and 1998, respectively. The 2000 net loss is before a $22.4 million deemed dividend related to the beneficial conversion of our preferred stock. Because we anticipate significant expenditures for our research and development programs and for the development, implementation and support of our gene function database, we expect to report substantial net losses through at least the next several years. We may never achieve profitability. If we do not become profitable within the time frame expected by securities analysts or investors, the market price of our stock will likely decline. If we do achieve profitability, we may not sustain or increase profitability in the future.

    We expect that our expenditures will continue to increase, due in part to:

•

continued investment in the research and development of our new and existing products and our technology, including increased investment for the development, implementation and support of our gene function database, our standard and conditional knockout programs and our secreted protein programs; and



•

our increasing investment in management and other employees, sales and marketing programs, customer service and operational and financial controls.

We are a newly public company with an unproven business strategy, and our limited history of operations makes evaluation of our business and prospects difficult.

    We have had a limited operating history and are at an early stage of development. Our strategy of offering a gene function database and using knockout mice to enable our customers to pursue promising candidates for drug target development is unproven. Additionally, our pricing models for offering our products and services are unproven. We currently have two subscribers for our gene function database. We have generated only limited revenues amounting to approximately $2.1 million, $1.2 million and $381,000 for the fiscal years ended 2000, 1999 and 1998, respectively. We recognized no revenue in 1997. Our success will depend upon, among other things, our ability to enter into licensing and other agreements on favorable terms, our ability to determine and generate information on those genes which have potential use as drug targets and the commercialization of products using our data. Moreover, we had no experience selling our data, and we had never provided a gene database before. Our sales force may not succeed in marketing our database product, and our employees may not succeed in implementing and operating our database in a manner that is satisfactory to our subscribers. Furthermore, the plans for our secreted protein and conditional knockout programs are unproven, and we cannot be sure that we will ever be able to develop these programs or that any program that we develop will be commercially successful. As a result of these factors, it is difficult to evaluate our prospects, and our future success is more uncertain than if we had a longer or more proven history of operations.

We currently have only four customers and will not succeed unless we can attract many more customers.

    We have only three customers for our DeltaSelect program. One of these customers is also a subscriber to our gene function database. We expect to enter into only a limited number of future DeltaSelect agreements. To succeed we must attract customers for our database and other programs. Our existing and future agreements may not be renewed and may be terminated without penalty in the

20

event either party fails to fulfill its obligations under one of these agreements. Failure to renew or the cancellation of these agreements by one of our customers could result in a significant loss of revenues. In 2000, GlaxoSmithKline and Schering-Plough accounted for 56% and 18%, respectively, of our revenues. Our current customers are GlaxoSmithKline, Pfizer, Schering-Plough and Merck.

    Over the past several years, companies in the pharmaceutical industry have undergone significant consolidation. If two or more of our present or future customers merge, we may not be able to receive the same fees under agreements with the combined entities that we were able to receive under agreements with these customers prior to their merger. Moreover, if one of our customers merges with an entity that is not a customer, the new combined entity may prematurely terminate our agreement. Any of these developments could harm our business or financial condition.

Because we have only recently begun to offer our database product, which we expect to be our principal source of revenue in the next few years, our future prospects are uncertain.

    We believe the majority of our revenues will be derived from fees under agreements with our database users. We may also derive revenues from royalties received from these users. We have entered into agreements with two subscribers to our database, however, we cannot be sure of the terms under which we may enter into future agreements, such as fees payable to us or the term of the agreements, if any. Our agreement with GlaxoSmithKline expires three years after our first delivery of data and provides for termination for any reason within the first three months after the one-year anniversary of the effective date upon the payment of a specified termination fee.

    If our database is not acceptable to our prospective customers, it may not generate revenues and our business and financial condition will be materially harmed.

    We may not be able to comply with minimum performance levels or restrictive provisions or other obligations that may be contained in any agreements, such as minimum data delivery requirements. In addition, we may experience unforeseen technical complications in the processes we use to generate functional data for our gene database and functional genomics resources. These complications could materially delay or limit the use of our gene function database, substantially increase the anticipated cost of generating data or prevent us from implementing our processes at appropriate quality and scale levels, thereby causing our business to suffer.

No drugs have been developed and commercialized using genomics-based research and therefore the future of our products and programs is uncertain.

    None of the limited number of drugs developed to date using genomics-based research have reached the commercial market. We cannot assure you that genomics-based drug development efforts will ultimately be successful. We have not proven our ability either to identify drug targets with commercial potential or commercialize drug targets that we do identify. We cannot assure you that a particular gene function in a mouse will have any correlation to a patient's response to a particular drug. It is difficult to successfully select those genes with the most potential for commercial development. Furthermore, we do not know that any products based on genes that are the subject of our research can be successfully developed or commercialized. If commercial opportunities are not realized from genomic-based research, our existing customers could stop using our products or we could have difficulty attracting or retaining customers and, in any event, we would not realize any product royalties.

21

Our customers will control the development and commercialization of products based on genes that we identify, which may mean that our research efforts will never result in any royalty payments or third party product sales.

    Our agreements with our customers may provide us with rights to obtain royalties from the commercial development of compounds or therapeutic approaches derived from access to our database, technology or intellectual property. However, we may not be able to obtain these rights under future agreements. Our ability to obtain these rights depends in part on the advantages and novelty of our technologies, the validity of our intellectual property, the usefulness of our data and our negotiating position relative to each potential customer.

    We will have limited or no control over the resources that any customer may devote to the development of compounds or therapeutic approaches derived from our access to our database. These customers may breach or terminate their agreements with us, and they are not obligated to conduct any product discovery, development or commercialization activities at all. Further, our customers may decide not to develop products arising out of our customer agreements or may not devote sufficient resources to the development, approval, manufacture, marketing or sale of these products. If any of these events occurs, our customers may not develop or commercialize any products based on our gene function research, technologies or intellectual property, we would not receive royalties on product sales and our results of operations would suffer. Furthermore, our customers may resist sharing revenue derived from the successful commercialization of a drug through royalty payments or others may have competing claims to all or a portion of such revenues.

There are a finite number of gene families upon which pharmaceutical and biotechnology companies focus their research, which limits our potential revenue and growth.

    Our current and potential subscribers and customers traditionally focus their research and development efforts on a finite number of gene families that they view as reliable drug targets. Once we provide functional information on these gene families, our ability to attract and retain subscribers to our database will depend, in part, on the willingness of our subscribers to expand their research and development activities to other gene families. If our customers do not do this, we may lose existing subscribers or fail to attract new subscribers for our database services, and as a result, our business and financial condition may be harmed. In addition, we have made and will continue to make significant investments in our database and knockout programs which we may not recoup if we cannot find additional target opportunities.

We may fail to meet market expectations because of fluctuations in our quarterly operating results, which could cause our stock price to decline.

    The following are among the factors that could cause our operating results to fluctuate significantly from period to period:

•

changes in the demand for and pricing of our products and services;



•

the nature, pricing and timing of other products and services provided by us or our competitors;



•

changes in the research and development budgets of our customers;



•

acquisition, licensing and other costs related to the expansion of our operations;



•

the timing of milestones, licensing and other payments under the terms of our customer agreements and agreements pursuant to which others license technology to us;



•

expenses related to, and the results of, patent filings and other proceedings relating to intellectual property rights; and

22

•

our unpredictable revenue sources as described below.

    We anticipate significant fixed expenses due in part to our need to continue to invest in product development and potential extensive support for our gene function database subscribers. We may be unable to adjust our expenditures if revenues in a particular period fail to meet our expectations, which would harm our operating results for that period.

    As a result of these fluctuations, we believe that period-to-period comparisons of our financial results will not necessarily be meaningful, and you should not rely on these comparisons as an indication of our future performance.

Our revenues will be unpredictable and this may harm our financial condition.

    The amount and timing of revenues that we may have from our business will be unpredictable because:

•

the timing of our DeltaSelect program agreements, gene function database subscriber agreements and installations are determined by our customers and subscribers;



•

whether any products are commercialized and generate royalty payments depends on the efforts, timing and willingness of our customers;



•

we do not expect to receive any milestone or royalty payment under licenses and other arrangements for a substantial period of time, if ever;



•

to date, we have entered into only two customer agreements for our gene function database and may not enter into any additional agreements;



•

the time required to complete custom orders for our database programs can vary significantly; and



•

our sales cycle is lengthy, as described below.

    As a result, our revenue may significantly vary from quarter to quarter, and our quarterly results may be below market expectations. If this happens, the price of our common stock may decline.

We expect that our sales cycle will be lengthy, which will cause our revenues to be unpredictable and our business to be difficult to manage.

    Our ability to identify and obtain subscribers for our gene function database product and other services depends upon whether customers believe that our products and services can help accelerate drug discovery efforts. Our sales cycle will be lengthy because of the need to educate potential customers and sell the benefits of our products and services to a variety of constituencies within potential subscriber companies. These companies are large organizations with many different layers and types of decision-makers. In addition, each database subscription and development program or services agreement will involve the negotiation of unique terms and issues which will take a significant amount of time. We may expend substantial funds and management effort with no assurance that a subscription program or services agreement will result. Actual or proposed mergers or acquisitions of our prospective customers may also affect the timing and progress of our sales efforts. Any of these developments could harm our business or financial condition.

We may have conflicts or be in competition with our customers, which will hurt our business prospects.

    We may pursue opportunities in fields, such as secreted proteins, that could conflict with those of our customers. Moreover, disagreements could arise with our customers or their partners over rights to our intellectual property or our rights to share in any of the future revenues of compounds or

23

therapeutic approaches developed by our customers. These kinds of disagreements could result in costly and time-consuming litigation and could have a negative impact on our relationship with existing customers. Any conflict with our customers could reduce our ability to attract additional customers or enter into future customer agreements. Some of our customers could also become competitors in the future. Our customers could develop competing products, preclude us from entering into agreements with their competitors or terminate their agreements with us prematurely.

We experience intense competition from other entities engaged in the study of genes, and this competition could adversely affect our business.

    The human and mouse genomes contain a finite number of genes. The human genome has been mapped and identified. Our competitors have identified and will continue to identify the sequence of numerous genes in order to obtain proprietary positions with respect to those genes. In addition, our competitors may seek to identify and determine the biological function of numerous genes in order to obtain intellectual property rights with respect to specific uses of these genes, and they may accomplish this before we do. We believe that the first company to determine the functions of commercially relevant genes or the commercially relevant portions of the genome will have a competitive advantage.

    A number of companies, institutions and government-financed entities are engaged in gene sequencing, gene discovery, gene expression analysis, gene function determination and other gene-related service businesses. Many of these companies, institutions and entities have greater financial and human resources than we do and have been conducting research longer than we have. In particular, a significant portion of this research is being conducted by private companies and under the international Human Genome Project, a multi-billion dollar program funded, in part, by the U.S. government, which completed and released its initial rough draft of the human genome in June of this year. Furthermore, other entities have and will continue to discover and establish a patent position in genes or gene sequences that we wish to study. Significant competition also arises from entities using standard target identification approaches, traditional knockout mouse technology and other functional genomics technologies. These competitors may have intellectual property rights in functional or other data which are superior to our rights. These competitors may also develop products earlier than we do, obtain regulatory approvals faster than we can and invent products and techniques that are more effective than ours. Furthermore, other methods for conducting functional genomics research may ultimately prove more advanced, in some or all respects, to the use of knockout mice. In addition, technologies more advanced than or superior to our gene trap technology and gene function identification technology may be developed, thereby rendering our gene trap and gene function identification technologies obsolete. As we expand our range of products and services, such as our secreted protein program, we will compete with additional companies, some of which may be our customers at that time or our potential customers.

    Some of our competitors have developed commercially available databases containing gene sequence, gene expression, gene function, genetic variation or other functional genomic information and are marketing or plan to market their data to pharmaceutical and biotechnology companies. Additional competitors may attempt to establish databases containing this information in the future. We expect that competition in our industry will continue to intensify. We also believe that some pharmaceutical and biotechnology companies are discussing the possibility of working together to discover the functions of genes and share gene function-related data among themselves. The formation of this type of consortium could reduce the prospective customer base for our gene function-related business. Moreover, the pharmaceutical industry has undergone significant mergers and this trend is expected to continue. This concentration of the industry could further limit our potential customer base and therefore harm our business.

24

If we fail to properly manage our growth, our business could be adversely affected.

    We expect to continue to experience significant growth in the number of our employees and the scope of our operations, including an increase in the scale of our mouse knockout program. As of December 31, 2000, we had 212 full-time employees. We expect our number of employees to continue to increase for the foreseeable future. In addition, we have substantially increased the scale of our knockout mouse production in the last year and expect to continue doing so for the foreseeable future. Our overall growth and need to develop many different areas of our company have placed, and may continue to place, a strain on our management and operations. If we are unable to manage our growth effectively, our losses could increase. The management of our growth will depend, among other things, upon our ability to broaden our management team and attract, hire, train and retain skilled employees. Our success will also depend on the ability of our officers and key employees to continue to implement and improve our operational and other systems. We will also be required to expend funds, which may be substantial, to improve our operational, financial and management controls, reporting systems and procedures.

    In addition, we will have to invest in additional customer support resources. Our current and potential database subscribers typically have worldwide operations and may require support at multiple U.S. and foreign sites and in multiple languages. To provide this support, we may need to open offices in addition to our Menlo Park and San Carlos, California facilities, which will result in additional burdens on our systems and resources and require additional capital expenditures.

We may engage in future acquisitions, which could adversely affect your investment in us as we may never realize any benefits from such acquisitions, which also could be expensive and time consuming.

    We intend to acquire and license additional products and programs, if we determine that these products or programs complement our existing technology or augment our existing information technology platforms. We currently have no commitments or agreements with respect to any material acquisitions. If we do undertake any transactions of this sort, the process of integrating an acquired business, technology, service or product may result in operating difficulties and expenditures and may absorb significant management attention that would otherwise be available for ongoing development of our business. Moreover, we may never realize the anticipated benefits of any acquisition. Future acquisitions could result in potentially dilutive issuances of equity securities, the incurrence of debt and contingent liabilities and amortization expenses related to goodwill and other intangible assets, which could adversely affect our results of operations and financial condition.

We may need to raise additional capital that may not be available, which could adversely affect our operations.

    Our products and services may not produce revenues that, together with our existing cash and other resources, are adequate to meet our cash needs. We plan to fund our operations from our existing cash balances, but we may in the future seek to raise additional funds from the sale of stock or from debt financing. Our cash requirements depend on numerous factors, including:

•

our ability to attract and retain customers for our gene function database and other products and services;



•

expenses in connection with the development and expansion of our gene function database, our secreted protein or other products and services;



•

expenditures in connection with license agreements and acquisitions of and investments in complementary technologies and businesses;

25

•

the need to increase research and development spending to keep up with competing technologies and market developments; and



•

expenditures as we expand our sales, marketing and customer service organizations and improve our management, operational and financial systems.

    If we need additional funding, we may be unable to obtain it on favorable terms, or at all. If adequate funds are not available, we may have to curtail operations significantly or obtain funds by entering into arrangements requiring us to relinquish rights to certain technologies, products or markets. In addition, if we raise funds by selling stock or convertible securities, our existing stockholders could suffer dilution.

We depend on key employees in a competitive market for skilled personnel, and without additional employees, we cannot grow or achieve profitability.

    We are highly dependent on the principal members of our management, operations and scientific staff, including William Matthews, Ph.D., our President and Chief Executive Officer, and Mark W. Moore, Ph.D., our Chief Scientific Officer. The loss of either of their services would harm our business.

    Our future success also will depend in part on the continued service of our key scientific, software, consultant and management personnel and our ability to identify, hire and retain additional personnel, including customer service, marketing and sales staff. We experience intense competition for qualified personnel. We may be unable to attract and retain personnel necessary for the development of our business. Moreover, our business is located in the San Francisco Bay Area of California, where demand for personnel with the skills we seek is extremely high and is likely to remain high. Because of this competition, our compensation costs may increase significantly.

We currently have no patents, and if we are unable to protect our proprietary information, our business will be adversely affected.

    Our business and competitive position depends upon our ability to protect and exploit our proprietary techniques, methods, compositions, inventions, database information and software technology. However, our strategy of obtaining such proprietary rights around as many genes as possible is unproven. Unauthorized parties may attempt to obtain and use information that we regard as proprietary. Although we intend for our gene function database subscription agreements to require our potential subscribers to control access to our database and information, policing unauthorized use of our database information and software may be difficult.

    We currently have no issued patents or registered copyrights. Patents have issued to other entities based on claims relating to knockout mice. In addition, many applications have been filed seeking to protect partial human gene sequences, many of which are based primarily on gene sequence information alone. Some of these applications have issued as patents. Some of these may claim sequences which we have used or may use in the future to generate knockout mice in our gene knockout program. In addition, other applications have been filed which seek to protect methods of using genes and gene expression products, some of which attempt to assign biological function to the DNA sequences based on laboratory experiments, computer predictions, mathematical algorithms and other methods. The issuance of these applications as patents will depend, in part, upon whether practical utility can be sufficiently established for the claimed sequences and whether sufficient correlation exists between the experimental results predictions, algorithms and other methods and actual functional utility. The patent application process before the United States Patent and Trademark Office and other similar agencies in other countries is confidential in nature. As each application is evaluated independently and confidentially, we cannot predict whether applications have been filed or which, if any, will ultimately issue as patents. However, it is probable that patents will be issued to our

26

competitors claiming knockout mice, partial human gene sequences and methods of using genes and gene expression products.

    Numerous applications have been filed by other entities claiming gene sequences. Many patents have already issued and we expect more will issue in the future. In addition, others may discover uses for genes or proteins other than uses covered in any patents issued to us, and these other uses may be separately patentable. We may not be able to obtain issued patents on our patent applications because our patent applications may not meet the requirements of the patent office. The holder of a patent covering a particular use of a gene or a protein, isolated gene sequence or deduced amino acid sequence could exclude us from using that gene, protein or sequence. In addition, a number of entities make gene information, techniques and methods publicly available, which may affect our ability to obtain patents.

    Some of our patent applications may claim compositions, methods or uses that may also be claimed in patent applications filed by others. In some or all of these applications, a determination of priority of inventorship may need to be decided in an interference proceeding before the U.S. Patent and Trademark Office. Regardless of determined outcome, this process is time-consuming and expensive.

    Issued patents may not provide commercially-meaningful protection against competitors. Other companies or institutions may challenge our or our customers' patents or independently develop similar products that could result in an interference proceeding in the U.S. Patent and Trademark Office or a legal action. In the event any researcher or institution infringes upon our or our customers' patent rights, enforcing these rights may be difficult and can be time-consuming. Others may be able to design around these patents or develop unique products providing effects similar to our products.

    Our ability to use our patent rights to limit com