Paul Pavlidis - US grants

Columbia University offers a wide range of training programs in the general area of Computational Biology. As has been the case for many universities, these programs have often resulted from the initiatives of individual faculty members or departments. In the past few years we have unified and consolidated these initiatives through the Center for Computational Biology and Bioinformatics (C2B2) and have created university-wide programs. These involve two campuses and a large number of departments. Our aims in the context of this proposal are to expand and improve what we have already created and to integrate the tools being developed in the MAGNet Center at all levels of our educational programs.

DESCRIPTION (provided by applicant): This proposal attempts to address the growing need for integrated gene expression informatics resources in neuroscience. The last five years have seen a rapid increase in the production of gene expression in all areas of biology. Analyzing this wealth of data is becoming more and more complex. While there are public data repositories where raw data can be published, there are few if any efforts to provide advanced analytic capabilities that address the specific needs of neuroscience. We propose to create such a facility that will allow neuroscientists to perform sophisticated computational analyses of large quantities of expression data coming from multiple laboratories. Data submitters will be able to control access to their data and the results coming from their data. The project will involve developing a database for meta-analysis of neuroscience-related gene expression data, tools for submission of data, and tools and algorithms for accessing the database and analyzing the data. Users will be able to perform meta-analyses of differential expression, coexpression (correlated expression), differential coexpression, expression detection, and of expression patterns in functionally-related groups of genes. In addition the system will incorporate other genomics data such as sequences, gene pathways and protein-protein interactions. By linking to and overlaying such data, users will be able to better predict the importance of particular analysis results. Finally, we will develop methods to incorporate domain-specific data into gene expression microarray analyses, such as neuroanatomy, neural connectivity and brain in situ hybridizations. This project will enable future collaborative efforts to target specific diseases or processes for examination.

DESCRIPTION (provided by applicant): An enduring challenge in biomedical research is deciphering the function of genes, and in particular how they work together to influence human health and disease. This project centers on the development and application of computational databases, tools and techniques for the study of large quantities of functional genomics data with a focus on the nervous system, building on our experience in meta-analysis of gene expression profiling data. Our first aim focuses on refining and applying methods for computational analysis of gene function in the nervous system, based on gene networks derived from expression profiling and other public data. Our second aim is to study the relationships between phenotypes and gene expression patterns, and applying the approaches to expression changes associated with diseases of the nervous system. Third, we propose to develop new visualization methods for gene networks, and to incorporate data on transcriptional gene regulation including transcription factor binding sites and genetic variation in gene expression. These resources will be designed to interoperate with other neuroinformatics databases, and disseminated through our "Gemma" web-based database system. PUBLIC HEALTH RELEVANCE: Disorders of the brain such as schizophrenia, autism spectrum disorder, Alzheimer's disease and stroke take a huge toll on society. Improving our understanding of how genes and gene networks contribute to normal and pathological processes in the brain will contribute to the development of improved diagnostics and treatments. This project will advance such understanding in multiple ways, by developing and applying computational analyses of huge quantities of genomics data on the brain.