IIT-D Develops New Bioinfo Apps

New Delhi | Updated: Mar 30 2004, 05:30am hrs
Researchers at the Indian Institute of Technology (IIT) Delhi have developed a comprehensive indigenous bioinformatic software called Sanjeevini for drug design. The researchers have also developed novel methodologies for protein structure prediction and genome analysis. The software and methodologies will help pharmaceutical companies and research and developments (R&D) units in drug discovery and development.

The development of these novel scientific methods and new software for genome analysis, protein structure prediction and active site directed drug design has been achieved at the supercomputing facility for bioinformatics and computational biology of IIT Delhi.

The supercomputing facility, created in 2002 with Rs 1.2 crore funding from the department of biotechnology (DBT), focuses on advancements in genomics, proteomics and drug design through research and training programmes and indigenous software development, IIT Delhi director RS Sirohi said.

He added that the facility, which has the best computing infrastructure in the bioinformatics learning space in the country, currently hosts an aggregate computing power of 150 billion operations per second and can store data up to 1.5 terabytes.

Giving details about the breakthrough in drug design, Prof Jayaram, coordinator, supercomputing facility, said the facility has developed a comprehensive indigenous active site-directed drug design protocol (software christened Sanjeevini 1.0) and tested it successfully on 65 inhibitors for COX-2 target enzyme, or to put it simply, proteins present in the human body. The protocol uses a supercomputer and methods of computational chemistry and biology and molecular biophysics to identify/design small molecules which preferentially bind to specific drug targets, he told eFE. Further tests on other drug targets are in progress and the facility is looking at the possibility of making the software commercially available at a later period.

Another achievement, according to Prof Jayaram, has been the development of a novel chemical model as opposed to the existing mathematical and statistical models for genome analysis.

The software named ChemGene 1.0 has been tested successfully on bacterial genomes and efforts to extend the methodology to eukaryotic genomes with eventual application to human genome are in progress. A genome sequence is only a first step towards identifying genes - regions of the genome that code for proteins - and understanding what each protein does in an organism.

Considered to be the holy grail of molecular biology and an outstanding problem for the last 70 years, any success in protein structure prediction has immediate applications to society, Prof Jayaram said. Structures for 10 small proteins have been predicted successfully with a methodology developed in-house and the target is to predict at least 100 proteins - the ubiquitous chemicals that govern virtually all of life processes in all organisms like bacteria, plants, humans and viruses - before the software is released. Understanding protein structures is fundamental to deciphering the biology of an organism which can help researchers find drugs or design vaccines.