Patrick O. Brown - US grants

The goal of the proposed work is to develop a radically new method for rapid direct mapping of the regions of genetic identity-by-descent between two related individuals, using only DNA samples from the two individuals. The method is intended to allow the simple and powerful but presently impractical "affected relative pair" mapping strategy to be put into practice. The principal advantages of the proposed method are: 1. No other family members need to be examined, and no pedigree information is required. 2. No locus-specific probes are required. 3. The entire genetic map is surveyed at high resolution in a single procedure, rather than through multiple discrete analyses of polymorphic loci at intervals throughout the genetic map. The method takes advantage of the fact that (at least in an outbred Western European or US population) any two allelic sequences that are not inherited from a common recent ancestor have single base differences on average every few hundred (about 300) basepairs. In other words, for an arbitrary pair of relatives, there are around 10-7 potentially informative "sequence polymorphisms" per haploid genome. The proposed method is designed to provide a practical way to exploit this vast source of information for gene mapping. The method involves four steps: 1. Preparing genomic DNA from two related individuals. 2. Preparing and isolating hybrid DNA molecules containing one strand from each individual. 3. Selecting those DNA hybrids that are free of mismatches over many thousands of base-pairs, and therefore very likely to represent sites of genetic identity by descent. 4. Using the resulting large, heterogeneous pool of perfectly, matched DNA hybrids as the template for probes, initially for in situ hybridization to metaphase chromosomes, to identify the map locations of sequences identical by descent between the two tested individuals. With a sufficiently large collection of affected relative pairs, the frequency of genetic concordance at the map position(s) of the trait-determining gene(s) will exceed that expected to occur by chance. In the case of a rare trait uniquely determined by a dominant allele of a single gene (e.g., von Recklinhausen's neurofibromatosis), 10 pairs of relatives sharing the trait would generally be sufficient to map the gene to a single 10 megabase region of the genome. Most of the technical obstacles to implementation of this approach have actually been overcome by previous workers, in other contexts. Thus, while the proposed concatenation and adaptation of available techniques to gene mapping represents a fundamental departure from previous mapping strategies, no fundamental technical breakthroughs are invoked. A closely-related method for mapping rare recessive traits using DNA from only one or a few affected individuals will also be tested.

Mapping genes that influence complex, multigenic or quantitative traits, traits that are relatively common in the population, heterogeneous in etiology,or that result from alleles with low penetrance, represents a principal challenge at the frontier of human genetics. In general, the solution to these genetic problems calls for a method that allows genotypes to be determined at very high resolution, for very large numbers of individuals, at a low cost in labor and materials. This proposal is directed at providing such a method, and making it widely available, convenient, and useful to the human genetics community. Genomic mismatch scanning (GMS) is a new approach to genetic linkage mapping, in which the regions of genetic identity-by-descent between two related individuals are mapped directly using only DNA samples from the two individuals. Each pair of relatives is analyzed in four steps: 1. Genomic DNA is prepared from each individual. 2. Hybrid DNA molecules, containing one strand from each individual, are prepared and isolated. 3. DNA hybrids that are free of mismatches over many thousands of base-pairs are separated from hybrids that contain mismatches. Since allelic sequences not inherited from a common recent ancestor have differences on average every few hundred basepairs, long mismatch-free hybrids are very likely to represent sites of genetic identity by descent. 4. The resulting pool of perfectly-matched DNA hybrids is labelled and used as a probe for in situ hybridization to metaphase chromosomes, or to an ordered array of cloned DNA's each covering a specific map interval. The contiguous intervals to which the probes hybridize correspond to regions of identity-by-descent between the two relatives. By collecting multiple pairs of relatives who share a trait of interest and, for each pair, mapping the regions where they share identity by descent, the genes that influence the trait can be mapped. Compared to methods currently in use, the principal advantages of GMS are: 1. The entire genetic map is surveyed at high resolution in a single, rapid procedure that does not require an electrophoresis step, resulting in a great reduction in cost and labor. 2. Discrete polymorphic markers for specific map intervals are not required but polymorphisms can be detected in virtually any arbitrary interval of interest. 3. Only "affected" family members need to be analyzed for linkage mapping. GMS can in principle also be used in linkage disequilibrium mapping using "unrelated" affected individuals to map a trait locus precisely. GMS has been tested successfully in Saccharomyces cerevisiae. The present proposal aims to develop, test and optimize GMS for widespread use in mapping human genes. The key elements of the proposal are: 1. Defining the optimal biochemical methods for GMS in humans. 2. Further developing the enzymatic and hybridization steps in the procedure. 3. Developing PCR- based variations on GMS for affected relative pair and linkage disequilibrium mapping in the human genome.

The early intracellular events in human immunodeficiency virus (HIV) infection, leading to the establishment of the integrated provirus, remain inadequately understood. We hypothesize that many steps in this process, presently unrecognized or unexplored, could provide new targets against which antiviral drugs might be developed. One of the better understood steps in early infection is integration. Integration has clearly been shown to be required for HIV replication, and the biochemistry of integration is understood in broad outline. Yet we are still far from a detailed understanding of integration, and consequently the current approach to developing drugs targeted at integrase may be naive. Indeed, consideration of the place integration occupies in the sequence of events leading to integration in vivo suggests that the most straightforward approach to blocking integration, based on inhibitors of the catalytic activities of integrase, may not be the most effective strategy for preventing its occurrence in an actual infection. The goal of the proposed work is simply to explore and develop the most promising avenues toward discovery of new antiviral agents directed in integration and other early steps in the infection process. Four projects are planned: 1. Continued basic investigations of the biochemistry of integration, aimed at developing a complete description of the regulation, specificity, biochemical mechanism and structure of integrase, and other possible actors in the process. As an integral part of this basic work we will continue to develop assays to test each activity that we can isolate, for possible use in primary or secondary screens of candidate drugs. 2. Exploring new strategies for antiviral agents directed at integrase and testing their feasibility by using mutations in integrase as surrogates for inhibitors. 3. Uncovering new targets for antiviral agents in early infection, using two genetic strategies. We will use a powerful new genetic approach to screen for new mutations in the gag and pol coding regions that impair specific steps in early infection. We will also continue work in progress, using the yeast two-hybrid system to screen for cellular proteins that interact with the Gag or Pol proteins of HIV or murine leukemia virus (MLV). 4. We will define the mechanism of action of integrase inhibitors identified by the ongoing random screens being conducted by the Parke-Davis group.

The megabases of DNA sequence that the genome project is amassing challenge us to discover the roles these sequences play in the life of an organism. This proposal aims to answer that challenge by using an efficient, sequence-based, genomic approach to construct a first- generation functional map of a complete eukaryotic genome. Investigations of the cell and molecular biology of Saccharomyces cerevisiae have made essential contributions to our basic understanding of how eukaryotic cells, including human cells, work. The sequence of well over half of the Saccharomyces genome has been determined, and the remainder is likely to be virtually completed within a year. Yet only about 25% of the putative genes that have been identified by sequencing the yeast genome have a function that is even partially understood. This proposal aims, in the next three years, to investigate the function of every gene in the Saccharomyces genome. Such an enterprise would be dauntingly expensive and laborious if conventional methods were used. An alternative approach, which exploits the economies of scale of a genome- wide approach, and uses the sequence as a resource, makes the proposed analysis practical and inexpensive. In the proposed "genetic footprinting" method of genome-scale functional analysis, a large library of new Ty1 insertion mutations will be generated by synchronously inducing transposition in a large population of genetically-homogeneous cells. Next, representative samples of the resulting mutagenized population will each be subjected to one of a set of at least 10 diverse selective conditions. Finally, each gene will be retrospectively screened for its effects on fitness under each selective condition. The retrospective screen employs the polymerase chain reaction (PCR) to determine whether cells carrying Ty1 insertions into a subject sequence were recovered following each specific selection. A role for the specified gene in the biological activity tested by a selection can be inferred from the relative depletion (or enrichment) of the cells carrying Ty1 insertions into that gene, when cell populations harvested before and after selection are compared. Because the most expensive and laborious steps - insertional mutagenesis and selections - are carried out only once for the entire genome, rather than separately for each gene, the proposed approach to determining the functions of genes is rapid and economical compared to those in current use. The feasibility of the genetic footprinting strategy has been tested extensively. To date, nearly 90% of the 266 putative genes on Chromosome V have been successfully tested for their effects on fitness under at least 4 selective conditions. These results establish that it is feasible to produce a functional map spanning the entire yeast genome, incorporating diverse tests for each gene's function, in 3 years, at a cost of about $220 per gene.

Mapping genes that influence complex, multigenic or quantitative traits, traits that are relatively common in the population, heterogeneous in etiology,or that result from alleles with low penetrance, represents a principal challenge at the frontier of human genetics. In general, the solution to these genetic problems calls for a method that allows genotypes to be determined at very high resolution, for very large numbers of individuals, at a low cost in labor and materials. This proposal is directed at providing such a method, and making it widely available, convenient, and useful to the human genetics community. Genomic mismatch scanning (GMS) is a new approach to genetic linkage mapping, in which the regions of genetic identity-by-descent between two related individuals are mapped directly using only DNA samples from the two individuals. Each pair of relatives is analyzed in four steps: 1. Genomic DNA is prepared from each individual. 2. Hybrid DNA molecules, containing one strand from each individual, are prepared and isolated. 3. DNA hybrids that are free of mismatches over many thousands of base-pairs are separated from hybrids that contain mismatches. Since allelic sequences not inherited from a common recent ancestor have differences on average every few hundred basepairs, long mismatch-free hybrids are very likely to represent sites of genetic identity by descent. 4. The resulting pool of perfectly-matched DNA hybrids is labelled and used as a probe for in situ hybridization to metaphase chromosomes, or to an ordered array of cloned DNA's each covering a specific map interval. The contiguous intervals to which the probes hybridize correspond to regions of identity-by-descent between the two relatives. By collecting multiple pairs of relatives who share a trait of interest and, for each pair, mapping the regions where they share identity by descent, the genes that influence the trait can be mapped. Compared to methods currently in use, the principal advantages of GMS are: 1. The entire genetic map is surveyed at high resolution in a single, rapid procedure that does not require an electrophoresis step, resulting in a great reduction in cost and labor. 2. Discrete polymorphic markers for specific map intervals are not required but polymorphisms can be detected in virtually any arbitrary interval of interest. 3. Only "affected" family members need to be analyzed for linkage mapping. GMS can in principle also be used in linkage disequilibrium mapping using "unrelated" affected individuals to map a trait locus precisely. GMS has been tested successfully in Saccharomyces cerevisiae. The present proposal aims to develop, test and optimize GMS for widespread use in mapping human genes. The key elements of the proposal are: 1. Defining the optimal biochemical methods for GMS in humans. 2. Further developing the enzymatic and hybridization steps in the procedure. 3. Developing PCR- based variations on GMS for affected relative pair and linkage disequilibrium mapping in the human genome.

Current systems for classification of cancer group together tumors with important differences in clinical behavior. As might have been expected from the manifest diversity in clinical behavior, we have found that there is enormous variation in gene expression patterns in tumors that would classically be grouped together. The variation among tumors in global gene expression patterns is, however, orderly and systematic and it provides a distinctive and reproducible signature for each patient s tumor, and even paints a picture of their biological differences. Moreover, we have found that variation in expression profiles can highlight unrecognized similarities and differences among tumors, and can provide a basis for systematic clustering of subsets of tumors. We therefore believe that underlying the apparent heterogeneity among cancers that we currently call by the same name, there may be a systematic taxonomy that is not readily apparent from histology or the small set of markers usually used to define subgroups of tumors. We propose to characterize the molecular variations among cancers of the breast, prostate, brain, and liver, by systematically and quantitatively measuring variation in transcript abundance for at least 20,000 different genes, in several hundred independent tumor samples from each of these tumor types. We will use multivariate clustering methods to search for ways to group tumors into clusters that are internally coherent in their expression patterns and thus, we hope, in their clinical behavior. Most of the tumor samples that are now available for the large retrospective studies that will be required to test the clinical utility of the new taxonomic groups we define are not suitable for analysis of gene expression at the RNA level. They are, however, well suited to immunohistochemical characterization. To make the transition from exploration and discovery of the molecular variation in cancer, to testing its connection to clinical behavior, we therefore propose to identify a large set of genes whose expression pattern varies most, and most independently, among the tumors we study, and raise antibodies against the predicted protein products. These antibodies will be used for immunohistochemistry, to characterize the variation in expression of the corresponding proteins among a diverse set of normal tissues, tumor samples and cultured cell lines. These antibody reagents will then be used for retrospective studies aimed at classifying tumors for which the natural history and treatment response is already known, to determine whether a new cancer taxonomy based on gene expression patterns can successfully order these cancers into groups with distinctive and consistent natural histories and patterns of response to treatment. These antibodies will aid investigations of the molecular pathogenesis of cancer. Some of them may provide a basis for non-invasive screens for early detection of cancers, and others could eventually even be used therapeutically.

From the start, the goal of this program has been the development and application of genome-wide approaches to systematic and quantitative analysis of gene expression patterns in cancer. In the eight years since we began this project, the experimental and analytical tools for systematic studies of global gene expression patterns at the level of transcript abundance have developed dramatically and they have become widely available and widely used for the study of cancer. The resulting studies have produced a wealth of new insight into cancer and they are even beginning to influence patient care. Profiling mRNA transcript levels, however, provides only a partial picture of the global gene expression program. While there is abundant evidence that regulation of the translation, subcellular localization and decay of mRNA are critical elements of biological regulation, practical and robust genome-wide approaches to studying these levels of regulation remain to be developed. As a result, our knowledge of the systems architecture, molecular mechanisms and biological roles of these regulatory mechanisms, including the roles they play in human cancer, is barely in its infancy. We therefore propose to develop practical, robust, high-throughput methods using DNA microarraysto profile three critical aspects of global regulation at the post-transcriptional level: translational regulation, regulation of mRNA degradation, and the specific interactions of RNA binding proteins with their targets. These studies will build on preliminary studies we've already begun in the Saccharomyces cerevisiae model system, but will focus on human cells. As the essential methodologies are developed, we will apply them to developing a foundational framework of knowledge, investigating the patterns in which these regulatory mechanisms are used in basic physiological and developmental programs, how they vary from one cell type to another and between individuals, and beginning to investigate the underlying molecular mechanisms. The goal is to develop both the experimental methodology and an interpretive framework to the point that these post-transcriptional levels of gene expression can be systematically profiled and studied on a genome-wide scale almost as routinely as transcript levels are today.

[unreadable] DESCRIPTION (provided by applicant):This program project seeks to improve the diagnosis and treatment of patients with cancer through a better understanding of cancer biology, improvements in cancer classification and diagnosis, and identification of new molecular targets and strategies for treatment. The multidisciplinary program emphasizes exploits systematic, highly parallel approaches to the molecular characterization of human cancer, particularly the use of DNA microarrays and tissue microarrays to profile gene and protein expression patterns. The central hypothesis is that any clinically important differences among cancers, or between cancers and their normal cellular counterparts, will be accompanied by a corresponding difference in gene expression programs. The research projects in this program include (1) a study aimed at identifying and characterizing the cells in the microenvironment of cancer tissues and their interactions, and; a systematic study of the cellular responses to molecular signals important in cancer; (2) a study of prostate cancer aimed at understanding the clinical implications and the biological basis of three distinct molecular subtypes of prostate cancer discovered in previous work, relating patterns of gene expression to chromosomal alterations, and developing and evaluating new prognostic markers; (3) a study of global gene expression and molecular diversity, in small, early stage breast cancers and a critical evaluation of the performance of gene-expression based prognostic criteria in predicting the clinical course of these early cancers; and (4) a large systematic investigation of the gene expression patterns in soft-tissue sarcomas, aimed at development of new molecular markers for diagnosis and new molecular targets for therapy of these cancers, and at improving our understanding of the stromal proliferation that is a common features of carcinomas. The program is supported by an administrative core (Core C) that will support, organize and provide supervision of the projects, and two research cores - a DNA microarray informatics and biostatistics core (Core A) that will support a microarray database for archiving, analysis and public distribution of data from this project and provide expert consultation and support for data analysis; and a tissue microarray core (Core B) that will produce and provide tissue microarrays, support for immunostaining and in situ hybridization, support for histopathological interpretation, and automated image analysis and image database support. [unreadable] [unreadable]

DESCRIPTION (provided by applicant): This proposal requests funds to purchase an Illumina Genome Analyzer II (GA II) as part of ongoing efforts to advance biomedical research at Stanford University. The proposed High-Throughput Sequencing (HTS) instrument will be placed in the Stanford Functional Genomics Facility (SFGF) within the School of Medicine and be a HTS resource for the Stanford Cancer Center. Since SFGF is a service center, the entire community of NIH funded researchers at Stanford will have access to the technology provide by the GA II. The facility presently offers expertise, support, and services for genomics and proteomics research utilizing microarray technologies. As many investigators are currently turning to HTS as the gold standard for functional genomics studies, it is critical that a shared resource be available so that the pioneering research pursued here continues. The HTS capabilities of the GA II will give NIH funded researchers at Stanford University the ability to further understand how our genetic code affects human health and disease. PUBLIC HEALTH RELEVANCE: The requested high-throughput sequencing (HTS) instrument will be used to monitor and investigate biological molecules utilized by the cell. Knowledge of how these molecules function will give us a better understanding of their role in human health and disease.