Title

A systems engineering approach for genetic pathway analysis and towards in-silico drug development

Date of Completion

January 2006

Keywords

Chemistry, Pharmaceutical|Engineering, Chemical|Biology, Bioinformatics

Degree

Ph.D.

Abstract

In this dissertation, we have proposed systems engineering approaches for improved gene expression profiling, genetic pathway analysis and drug development pipelines. The models for genetic pathway analysis consist of (a) hybrid Boolean algebra/network concepts, and (b) non-linear programming (NLP) models. We have shown that the hybrid network model can effectively simulate natural biological processes and model the gene regulatory systems. The study investigated the time evolution of gene expression and the results from the model compared very well with the experimental data. The suggested model is better in the sense that it allows for continuous inputs and captures the inherent rate process (via for example time delay and rate constants). In addition it reveals the underlying genetic pathways and provides valuable biological insights. The proposed model would be appropriate in situations where either enough experimental data is not available or experiments cannot be performed. Central to the network model is the need to cluster the genes based on their expression levels. To do this efficiently we employ the ACA (Adaptive Centroid Algorithm, clustering algorithm) appropriately modified to include significant statistics. ^ In the second part of the dissertation, we investigated the molecular recognition of active sites as a rational approach towards pharmaceutical drug design Here we employed computational chemistry and group contribution theory. We assumed that the structure of the target protein was known a priori, which forms the basis for investigating the structure based drug discovery. Considering that biological activity and selectivity are the most important attributes for a therapeutic drug, the suggested procedure can be used (a) for virtual screening of drugs in the final stages of pre-clinical development and (b) for the design of new or improved drugs. It has been successfully demonstrated that the chances of a drug reaching the market in a timely fashion are enhanced, while the likelihood of drug failures is minimized (with the development of better COX-II inhibitors). ^ In summary the dissertation suggests how systems engineering (exploiting optimization, statistics, gene expression biology, and quantum mechanics) can be employed towards genetic pathways and structure based drug design. ^