Abha Singh Bais

Research Interests:

Until now, my research interests lay at the intersection of comparative genomics and transcriptional regulation. In broad terms, various algorithmic and statistical problems in molecular biology interest me. These include, but are not restricted to: gene regulation, alignments, evolutionary models, chromatin structure, hidden markov models, Markov Chain Monte Carlo methods, DNA segmentation, etc.

My PhD research can be categorised into two main themes:

• regulatory element analysis: probabilistic models for transcription factor binding sites (TFBSs), conservation and evolutionary characteristics of TFBSs, cis-regulatory modules, interaction/competition between TFBSs, etc.
• pairwise alignments: algorithms for optimal and sub-optimal alignments, issues related to parameter choice, interplay with transcription factor binding sites.

Briefly, in my PhD, I developed an algorithm for the prediction of conserved transcription factor binding sites during the generation of alignments. For parameter choice, I proposed a statistical framework which does not require training. In Hidden Markov Model (HMM) terminologies, the algorithm can be formulated as a pair HMM with additional states for every new profile (position-specific probability matrix) to be searched for. Later, I extended the approach to allow for the possibility of considering evolutionarily related binding sites. Both the tools SimAnn and eSimAnn are available for download (see below). If you are interested to know more, please see [abstract | thesis].

Short Vita: