Rajesh Nair, PhD
Associate Research Scientist
Center for Computational Biology and Bioinformatics (C2B2)
North Eastern Structural Genomics Consortium (NESG)
Background
A theoretical physicist by training, I was drawn to Biology by the complexity and diversity of problems that could be addressed using quantitative approaches. The primary focus of my research has been developing machine learning models that can help us better understand the structure and function of proteins. My doctoral thesis at Columbia University was on the computational prediction of protein subcellular localization, an important aspect of the function of proteins.
Research Interests
My recent research has focused on strategies for identifying good protein targets for structure determination at NESG, which is one of four large centers funded by the protein structure initiative (PSI). Target selection plays a vital role in high throughput structure determination by improving success rates and leading to structures with novel folds and higher modeling leverage.
My primary research interest is developing machine learning models to explain real world scenarios. One problem I have been working on recently is developing machine learning models that can rapidly integrate and adapt to the volume of high throughput functional information being generated in molecular biology. Another area of research is developing models for automatic motif detection.
Education
I did my undergraduate studies at the Indian Institute of Technology (IIT) at Kharagpur in Physics and Math, graduating in 1997. Following this I joined the Physics department at Columbia University, where I started working on string theory. However, I found myself excited about the prospect of developing quantitative models to explain biological problems and so I switched my research focus to Bioinformatics. I obtained my Ph.D. in Physics from Columbia University in 2004; the title of my dissertation was Computational approaches for predicting the subcellular localization of proteins (read thesis), and Burkhard Rost was my advisor.
WWW Services:
Below are quick links to some popular services. Descriptions and links to additional services may be found under the Services tab above.
 |
LOCtree
LOCtree is a novel system of support vector machines (SVMs) that predict the subcellular localization of proteins, and DNA-binding propensity for nuclear proteins, by incorporating a hierarchical ontology of localization classes modeled onto biological processing pathways.
>> more information |
| |
|
 |
PredictNLS
PredictNLS is an automated tool for the analysis and in silico determination of Nuclear Localization Signals (NLS).
>> more information |
|
