Aditya Pratapa
I am an MS Scholar at the Computational Systems Biology lab at IIT Madras. I’m currently working on designing efficient algorithms to evaluate the knockout behavior in organisms using genome-scale metabolic models. The study is aimed at predicting metabolic capabilities of organisms using flux balance analysis (FBA), a simulation tool based on Linear Programming for the in silico analysis of reconstructed metabolic models. To understand the biological robustness, simulations are performed to understand the knockout behaviour. These simulations are performed by combinatorially removing variables in the linear optimization problem that would result in simultaneous restriction of the maximum value of objective function to zero, while satisfying the linear mass balance constraints. This would potentially involve solving a few hundred million to few billion LP problems as a result of combinatorial explosion. Towards this end, I developed an algorithm, FastSL, to predict the knockout behaviour, that reduces search space over 4000 fold and is at least 40 times faster than existing methods. MATLAB implementation compatible with COBRA Toolbox is available for download at GitHub.
Publication(s):
- Aditya Pratapa, Shankar Balachandran and Karthik Raman.Fast-SL: An efficient algorithm to identify synthetic lethal sets in metabolic networks. Bioinformatics. doi:10.1093/bioinformatics/btv352
Contact: adityapratapa@smail.iitm.ac.in