Our goal is to accelerate the rate of discovery of novel therapeutic agents by rapidly assigning function to genes with no experimental functional data (NEFD) and organizing them within Protein Interaction Network (PIN) maps. This will facilitate the identification of novel candidate drug targets in disease-associated pathways, such as osteoporosis and Alzheimer's Disease. To begin with, we have mined the public domain and identified a sub-set of NEFD genes (1241) to populate a prototype knowledge database. The design of the database is based on a generic data model with a web based user interface.
The aim i s to build an Integrated Knowledge Workbench (IKW) in which all data, determined in silico and experimentally, will be incorporated. To transition the prototype to the IKW, we will integrate a suite of bioinformatics analysis tools that will be used to generate detailed in silico fingerprints for every NEFD gene. Furthermore, in Phase I we will integrate protein interaction data derived from our yeast two-hybrid experimental platform. Results from these analyses will be incorporated systematically into the knowledge-based component of the database. To test the platform we will randomly select 5 NEFD genes, generate in silico fingerprints and identify their protein interactors. These data will be integrated to assign a function and produce PIN maps that will chart each of the NEFD genes in a pathway. This test of the system will illustrate the utility of the integrated platform to obtain pertinent information about genes of unknown function and will begin to assist in the determination of their potential role in osteoporosis and Alzheimer's disease-associated pathways.
| Wang, Huajun; Huang, Hui (2004) SED, a normalization free method for DNA microarray data analysis. BMC Bioinformatics 5:121 |
