ABI Development: Gene function visualization and inference across all plants

Phoenix Bioinformatics Corporation, Fremont CA

Investigators

Abstract

Addressing the food needs of an increasing global population in the 21st century will require significant improvements in agriculture akin to the Green Revolution of the previous century. To address this goal it is of critical importance to understand as much as possible about the functions of the tens of thousands of genes present in each plant species, in order to select those functions of most relevance to engineering favorable traits such as drought tolerance or increased grain quality. Much of what is currently known about plant gene function comes from experiments in model plant systems such as Arabidopsis thaliana, but because of evolutionary relatedness this information can be used to predict the functions of genes in crop species such as maize and rice, and to provide targeted hypotheses that can be followed up by further experiments. This project will increase both the efficiency and accuracy of gene function prediction in plants by organizing and integrating high quality data from a number of sources and presenting it in a user-friendly format that makes clear the evolutionary relationships between genes. The team will develop software for visualizing the evolutionary relationships as a gene family tree, providing an organizing framework for disparate experimental observations from a number of experimental systems. The project will provide software for use, and for data sharing, by a broad community of both researchers and students. The project?s inclusion of local community college students will provide training and increase participation in computer science, genomics and plant biology by groups that are underrepresented in these fields. The project output will be a powerful and intuitive software tool for inferring plant gene function from phylogenetic relationships and experimental evidence. To achieve this goal the project aims are to: 1) assemble sets of phylogenetic trees using selected model organism and plant genomes, along with a suite of web services to ensure easy access to the data; 2) follow user-driven design principles to design and build the user interface, in order to visualize the phylogenetic trees and associate the genes within each tree with high quality functional information including Gene Ontology annotations, gene expression patterns, protein domains, sequence variations and mutant phenotypes; 3) provide an analysis platform that streamlines workflows and includes personal workspaces for users to add their own data and make it public when their work is completed, thus enriching the knowledgebase; and 4) produce workshops, course materials and online content to train scientists at all academic levels on how to use the software. The results of the project will be available at http://www.phoenixbioinformatics.org/treevision/.