I am a computational biologist researching at Merck Research Laboratories (MRL). I use computational methods to generate and validate testable hypotheses that accelerate data-driven discovery. Prior to MRL, I was a postdoctoral fellow of Computational Biology and Bioinformatics at Harvard and a PhD candidate of Biostatistics at UAB.
Postdoctoral Fellowship in Computational Biology and Bioinformatics, 2019
Harvard University and Broad Institute
PhD in Biostatistics, 2015
University of Alabama at Birmingham
MSc in Statistics, 2009
Indian Institute of Technology Kanpur
BSc in Statistics, 2007
University of Calcutta
Statistical computing for reproducible research
Exploratory data analysis and visualization
Version control for scientific workflows
Development and deployment of ML models
Scalable Bayes and uncertainty quantification
Analysis and interpretation of biological data
Led multiple data science projects in the preclinical Biostatistics space as part of the Early Development Statistics within the global Biostatistics and Research Decision Sciences (BARDS) organization.
Employed machine learning classification to the discovery of clinically actionable biomarkers from integrated multi-omics profiles to enable better disease outcome prediction and patient stratification.
Published research across several disease areas (infectious diseases, oncology, and microbiology) that utilized innovative statistical methodology and systems biology techniques spanning a wide range of translational applications.
Developed a computational method MelonnPan to predict metabolite profiles from metagenomic data using concepts from machine learning and ecology, implemented in R/Bioconductor.
Contributed to the development of MaAsLin2, an R/Bioconductor package for associating microbial multi-omics data with arbitrarily complex clinical metadata using linear models.
Contributed to grant writing, manuscript preparation, interdisciplinary collaborations, and teaching and mentoring of graduate and undergraduate students and trainees.
Developed Bayesian machine learning methods for high-dimensional feature selection in personalized medicine applications.
Developed and validated risk prediction models for assessing short-term mortality in obese adults.
Conducted high-dimensional predictive modeling of zero-inflated count phenotypes to identify genetic susceptibility markers in Rheumatoid Arthritis patients.