Analysis of gene expression in human cancers and cancer models
This work, carried out in collaboration with Timothy J. Yeatman (http://www.moffitt.usf.edu/Prevention_and_Treatment/physician/details.asp) of the H. Lee Moffitt Cancer Center in Tampa, Florida, seeks to use DNA microarray analysis to understand the mechanistic basis of human colon and breast cancers and to develop diagnostic and prognostic ?fingerprints? that can help in the detection and management of the disease. Our work combines the use of microarrays to monitor changes in gene expression and to detect changes in the structure and organization of the human genome with advanced computational techniques to integrate data from a wide range of other sources in an attempt to better understand how multiple factors combine to influence the development and outcome of the disease. With candidate genes in hand, we hope to understand the mechanisms responsible by perturbing the expression of these genes and reconstructing the pathways and networks in which they participate.
The Gene Index Databases
The promise of genome projects has been a complete catalogue of genes in a wide range or organisms. While genome projects have been successful in providing reference genome sequences, the problem of finding genes and their variants in genomic sequence remains an ongoing challenge. The sequencing of Expressed Sequence Transcripts (ESTs), fragments of genes that have been copied from DNA to RNA, provides the most comprehensive evidence for the existence of genes and their structure. The goal of The Gene Index Project is to use the available EST and gene sequences, along with the reference genomes wherever available, to provide an inventory of likely genes and their variants and to annotate these with information regarding the functional roles played by these genes and their products. In addition, we are attempting to use these catalogues to find links between gene genes and pathways in different species and to provide lists of features within completed genomes that can aid in the understanding of how gene expression is regulated.
Software tools for `omics data
DNA microarrays allow us to look at patterns of gene expression for tens of thousands of genes in a single assay. While this gives us unprecedented power to examine the way in which genes can influence an organism?s ability to respond to its environment, it also presents unprecedented challenges with respect to the collection, management, analysis, visualization, and interpretation of the data. While genome projects have managed larger quantities of data, the complexity of analyzing expression patterns is significantly greater as we must not only understand what genes are represented in any assay, but we must also have detailed information about the state of the biological samples under analysis. Our goal has been to create a collection of freely-available, open-source, user-friendly software tools for the analysis of these ?higher dimensional? datasets. The success of these tools is driven by their focus on addressing biological problems encountered in the analysis of real data and through a close collaboration between computational and laboratory scientists.
See the software developed by our group on the TM4 web site.