Comparative Genomics-driven Discovery of Maize Metabolic Functions

点击翻译按钮获取中文摘要

PI: Andrew D. Hanson (University of Florida - Gainesville)CoPIs: Valérie de Crécy-Lagard, Donald R. McCarty, and Jesse F. Gregory (University of Florida - Gainesville), Christopher S. Henry (University of Chicago)Senior collaborators: Andrei Osterman (Burnham Institute for Medical Research) and Svetlana Gerdes (Fellowship for Interpretation of Genomes) The rapid progress in sequencing plant genomes and genes has exposed the lack of information regarding the function of 70% of the proteins encoded. In addition, these studies have shown that many such unknowns also occur in bacteria and archaea. Comparative genomics is a powerful approach to uncover gene function, as is advanced in silico reconstruction of an organism's metabolism. In this project, both approaches will be used to improve the accuracy of functions assigned to maize metabolic genes and to predict functions for unknown genes, with special emphasis on B vitamins (folate, niacin, thiamin, pyridoxine, riboflavin, pantothenate, and biotin). The ten most promising functional predictions will be experimentally validated by combining genetic and metabolic profiling approaches in bacteria with biochemical assays of recombinant proteins and with genetic tests in maize. The expected overall outcome of this project is the implementation of a comparative genomics prediction and validation pipeline for maize gene function discovery, using B vitamin metabolism as a paradigm. Gene functions (annotations) will be publicly available via MaizeCyc (http://pathway-dev.gramene.org/gramene/maizecyc.shtml) and the SEED (http://theseed.uchicago.edu/FIG/index.cgi) databases, with metabolic reconstructions available via the Model SEED website (www.theseed.org/models/).The project will implement an interdisciplinary approach to gene function discovery that is extendable to any metabolic network. In the process, it will enrich the maize genome by imposing consistency on thousands of metabolic gene annotations, and by improving the annotations of hundreds of unknown genes. Both impacts will advance future function discovery. Furthermore, by developing genome-scale in silico metabolic reconstructions, the work will initiate a systems approach to understanding maize metabolism. Integrally, the work will provide cross-disciplinary training in comparative genomics, metabolic biochemistry, and microbial genetics to students, postdoctoral associates, and faculty. In addition the project will provide for an annual 3-day hands-on workshop at the University of Florida to train researchers at all levels to predict functions using the SEED and other comparative genomics databases, with special emphasis on training faculty from minority serving institutions. Finally the project will develop an undergraduate bioinformatics course in which the students participate in unknown gene identification and metabolic reconstructions using comparative genomics and a graduate course module in which students develop functional predictions for unknowns in the project.