Chemical-Bioinformatics using Machine Learning Techniques

Identification of potential drug molecules (known as lead discovery and optimization) accounts for about one-third of the drug development cost and considerable amount of time. Pharmaceutical companies are therefore investing significant resources in developing novel computational techniques to reduce the time and cost early in the drug development process. The objective of my research is to develop novel computational models and software for bioactivity prediction and data mining of drug molecules from large databases. Using supervised machine learning techniques (i.e., GA and PCA for classification and feature selection followed by ANN or SVM techniques for pattern recognition), in combination with statistical regression analysis, we are currently developing robust computational models for rapid and reliable prediction of biologicalactivity of HIV protease inhibitors (i.e., potential HIV drug compounds). This research is being carried out in collaboration with Prof. R. Garg who has research grants from NIH and industry. 

Projects:

• Hybrid GA-ANN based Computational Prediction Tools for Drug Des
• Shape Analysis Techniques for HIV Protein -Ligand Interactions