Functional Annotation of Protein Interactome Graphs

点击翻译按钮获取中文摘要

Our project aims at understanding metabolic disorders such as diabetes and insulin resistence by means of a phyletic comparison between different organisms since metabolic pathways are highly conserved between species. Creating the bipartite graph requires a database of the different biomolecules and their reactions. After integrating biomolecule data into the database, enzyme-substrate, protein-protein, and cofactor-enzyme reactions are required. Currently, our source for this information is the Kyoto Encyclopedia of Genes and Genomes (KEGG). Since biological data sources develop rapidly and new data sources appear very often, extensibility is extremely important to the data structure of the program. We therefore decided to create our own XML schema to store the data. A client-server software design enables features that would be hard to do otherwise with a browser-embedded software design. Additionally, the modular design was adapted to be extensible so that other scientists are able to add features/data to the existing data structure with ease. The software is designed primarily to database and display the human interactome in a bipartite manner. Additional capabilities however, are also currently being designed into the program structure. Some of these abilities are: (1) The ability to display a series of reaction (shortest path or otherwise) from one molecule (either a compound or protein), to another molecule, (2) The ability to demonstrate a signals propagation along a pathway upon perturbation of a constant (either a molecular concentration or reaction kinetic), (3) The ability to selectively create putative reactions as part of the normal interactome dataset, and (4) The ability to do a phyletic comparison of pathways between different organisms to enable the utilization of experimental data from different organisms in mathematical model building.