RNA Structure Estimation from Multiple Homologs[project page]

Nature encodes the functional structure of a (noncoding) RNA molecule as multiple alternative homologous sequences in different organisms. In our work, we seek to decode this common structure by exploiting this comparative information across the multiple homologs in combination with thermodynamic models for predicting secondary structure.

Data Analytics for Large Scale Flow Cytometry Data[project page]

Flow cytometry (FC) is a powerful technology for rapid measurements of multiple attributes at the individual cell level for populations with millions of cells. In our work, we focus on automated analysis of the large scale datasets generated by flow-cytometry.

Sensor Data Analytics for Parkinson’s and Huntington’s Diseases 

Parkinson’s and Huntington’s diseases are progressive neurological conditions that cause movement disorders in affected individuals. In our work we utilize multiple body-affixed sensors to obtain long duration three dimensional accelerometer data. We use machine learning techniques on the data for tracking progression of disease, medication efficacy, enabling personalized medicine and early prognosis.

Selected Publications