GOVERNMENT

SRI International, an independent nonprofit research and development organization, has performed a computational analysis of the human genome to predict metabolic pathways and to predict new gene functions within the human genome. Using SRI's PathoLogic software, the analysis assigned 622 human enzymes to roles in 135 predicted metabolic pathways. The pathways and the analysis results are available in the HumanCyc database at humancyc.org. The HumanCyc database is the seventeenth in SRI's growing collection of BioCyc (http://biocyc.org/) pathway and genome databases. HumanCyc provides a genome-based view of human nutrition that could foster a better understanding of the links between genome, diet and health. "The human genome is incomprehensibly vast and complex; Pathways provide a framework for organizing the human genome so that scientists can more easily understand and manipulate it according to molecular interactions," said Peter Karp, Ph.D., director of SRI's Bioinformatics Research Group. "SRI's goal is to provide biologists with the 'power tools' they need to understand and analyze the genome in a much more useful way. By structuring information into pathways, we provide a unique genomic framework to more easily group and analyze our biochemical machinery. For example, one pathway in the HumanCyc database describes genes that work together to break down nicotine; another pathway contains the group of genes that make cholesterol." One of SRI's software tools, the HumanCyc Omics Viewer, allows researchers to visualize combinations of gene expression, proteomics, and metabolomics data by painting them onto the cellular overview of human metabolism. This pathway framework also allowed SRI to identify 203 probable missing enzymes in the human genome, and SRI's newly developed pathway hole filling algorithm generated high-scoring candidate genes for 25 of those enzymes. As part of this work, SRI researchers compared the metabolic pathways of humans and two other well-understood organisms, Escherichia coli (the bacteria better known as E. coli) and Arabidopsis thaliana, a small flowering plant widely used as a model organism in plant biology. The researchers were able to identify 35 pathways shared among the three organisms: human, bacteria and plant. For this work, SRI researchers partnered with researchers at Stanford University and Indiana University-Purdue University of Indianapolis. An article about the group's collective findings, "Computational prediction of human metabolic pathways from the complete human genome," was published in the December 22, 2004 issue of Genome Biology.