My research interests are design and analysis of algorithms and combinatorial optimization with applications in computational molecular biology.
As a Ph.D. student, I had worked on topics in sequence analysis including local alignment by maximum surprise and inverse parametric sequence alignment. The local alignment by maximum surprise is a problem to find the most surprising local alignment between two input strings that has the least probability of occurring by chance. I studied a nonlinear similarity function for this problem and derived an accurate analytic approximation to this probability function that generalizes known special cases. The inverse parametric sequence alignment is a problem to learn optimal parameter values for the sequence alignment objective function from input biological reference alignments, or simply examples. I designed and implemented efficient algorithms for this problem, and developed general alignment models for protein sequences that incorporate predicted secondary structure in the examples to further improve alignment accuracy.
As a postdoctoral researcher, I am involved in a project called whole genome alignment and working on a problem of finding homologous substrings in DNA sequences and building a phylogenetic tree with them.