About me:

A computer scientist, with research domain in computational biology, epigenetics, and human genetics, and focus in developing novel algorithms, statistical and AI/ML models, and bioinformatic frameworks utilizing multi-omics NGS data. 

Research objective: 

1. What are the functional genes/drug targets/gene loci, regulatory elements, and genetic variants driving context-specific (cell-type, disease-specific, or condition-dependent) genetic and epigenetic regulation?

2. And, how to derive such targets by developing novel computational algorithms, AI/ML, deep learning, statistical models, and bioinformatic frameworks on multi-omics NGS data (transcriptomics, epigenomics, 3D chromatin, human genomics, single cell and multi-omics)?

Research accomplishments so far in a nutshell: 

1. Developed novel computational algorithms and statistical models for identifying epigenetic regulators from 3D chromatin looping data (Nature Communications 2019; Nature Protocol 2020; Cell Reports Methods 2025; Genome Biology 2025)

2. Developed graph-based deep learning and explainable AI to predict gene expression and epigenetic regulators (bioRxiv 2026).

3. Developed statistical models to derive putative functional genetic variants (SNPs / QTLs) from eQTLs, epigenetic data, and chromatin interactions (Nature Communications 2024; Nature Genetics 2021; Science Immunology 2022)

4. Identified functional genetic variants, cell-type-specific gene regulation, and potential disease-specific loci: 1) using COVID-19 GWAS SNPs (Nature Communications 2021), 2) epigenome and 3D chromatin loops for multiple myeloma (Nature Communications 2019) and B cell differentiation (Cell 2023).

5. Utilized single cell RNA-seq and multi-omics data to: 1) Identify marker genes and cell types for skin disorders (in preparation), 2) identify cell subsets for anti-PD-1 therapy for lung cancer (JEM 2019).

6. Developed XGBoost model for missense proteome variant classification, and also developed GWAS inference frameworks (PLINK, SAIGE, fastGWA, REGENIE) for common and rare variant identification (Poster, ASHG 2025).

7. Worked on various NGS datasets: 1) short and long-read RNA-seq (PacBio, ONT), 2) Single cell RNA-seq, CITE-seq, multi-omics (scRNA-seq + scATAC-seq), 3) Epigenomics: ChIP-seq, ATAC-seq, multi-omics, methylation, 4) Chromatin interactions (Hi-C, HiChIP), 5) Human genetics: GWAS, QTL (eQTL, caQTL), 6) Proteomics: missense mutations.

Future research directions:

1. Using deep learning and graph-based models to prioritize drug targets by integrating above-mentioned genomic and epigenomic signatures.

2. Integrating single cell and multi-omics perturbation signatures with genomic and epigenomic evidences in deep learning frameworks to pinpoint causal genes.

Career

Lead Data Scientist (October 2024 - ), Empirico,  San Diego, CA 92130, USA

1. Developed scalable GWAS inference frameworks (PLINK, SAIGE, fastGWA, REGENIE) and benchmarked for both binary and quantitative traits.

2. Developed novel statistical and matrix decomposition methods to achieve ∼6X speedup on GWAS null model inference and ∼100X speedup on GWAS association test - applied on UKBB and Genomics England datasets.

3. Developed long-read transcriptomics pipelines (PacBio, ONT) to discover novel isoforms with disease relevance.

4. Designed ML algorithms for classifying missense protein mutations, aiding variant pathogenicity prediction.

Instructor (January 2021 - October 2024), La Jolla Institute for Immunology, La Jolla, CA 92037, USA

1. Developed deep learning model (CNN + GNN) to predict gene expression from epigenomic and 3D chromatin looping.

2. Developed computational methods to derive putative functional SNPs: 

   1. novel QTLs from chromatin interactions 

   2. eQTLs from scRNA-seq 

   3. causal GWAS SNPs using transcriptomic and epigenetic data for COVID-19 and Type 1 Diabetes.

3. Developed method to call differential 3D chromatin loops and used them to model 3D chromatin changes in B cell differentiation.

4. Mentored researchers and grad students for: 

   1. single cell and multi-omics analysis for various immune diseases and skin disorders 

   2. design a database for storing HiChIP loops and SNP-to-gene links.

Postdoctoral researcher (April 2017 - December 2020), La Jolla Institute for Immunology, La Jolla, CA 92037, USA

1. Developed computational methods to identify regulatory 3D chromatin (HiChIP and Hi-C) interactions. 

2. Developed novel statistical and graph-based algorithm to identify functional eQTLs using chromatin interactions and epigenetic data.

3. Analyzed 3D regulatory changes in cancer (multiple myeloma) patients with 4:14 translocation.

4. Collaborated with immunologists to identify immune cell subsets related to cancer therapy using scRNA-seq.

Ph.D. (July 2012 - August 2017), Computer Science and Engineering, Indian Institute of Technology Kharagpur, India

1. Thesis topic: developing algorithms in Computational phylogenetics and molecular evolution. 

1. • Resolving the conflict of evolutionary information among a group of species, due to the difference in individual genetic lineages.

Master of Science (M.S.) by research + Project Scholar (July 2009 - July 2012), Computer Science and Engineering, Indian Institute of Technology Kharagpur, India

1. Thesis topic: Machine learning assisted detection of epileptic seizure patterns using neonatal electroencephalogram (EEG) signal and video recordings.

Research Software Engineer in STMicroelectronics, India (July 2006 - July 2009) (R&D Section)

1. Video transcoding between formats H.264, MPEG-4, and MPEG-2, along with the bit rate control mechanism, for supporting low bandwidth.

   • Integrated in STMicroelectronics implemented DBS (Dynamic Bitstream Shaper) transcoding library (integrated into STm71xx product family).

   • Integration of A/V codecs in  OpenMAX (Open Media Acceleration) and MPLAYER (version 1.0pre8) multimedia libraries.

Publications

See this page and also Google Scholar

GitHub repositories

Check out this page

Talks

See this page

Skills

Check out this page for details.

Research Grants

• NIH R03 grant 1R03OD034494-01 (Using Common Fund datasets for prioritization of disease-associated genetic variants): Assisted my supervisor Dr. Ferhat Ay in preparing the grant proposal, initial results. (link)

• D - Challenge 2021: Our team obtained a grant of $20,000 for research in Type 1 Diabetes (T1D), from SugarScience (https://thesugarscience.org/). For details, check http://info.thesugarscience.org/dknet-21-d-challenge

News articles:

• MARCH 2022: Our work on single-cell eQTL in immune cells is featured in various scientific news articles.

• NOVEMBER 2021: Our work on Functional variants of COVID-19 in immune cells is featured in various scientific news articles.

• JANUARY 2021: Our work on promoter interacting eQTL is highlighted in various scientific news articles.

• JUNE 2019: Our work Single-cell transcriptomic analysis of tissue-resident memory T cells in human lung cancer is featured in various scientific news articles. 

Awards / Fellowship / Travel Grants 

• JULY 2022: Conference travel fellowship from ISMB 2022.

• JUNE 2015: Conference Travel fellowship from the Organizing Committee of ISBRA 2015.

• MAY 2015: Conference Travel fellowship from the Department of Science and Technology (DST), India, for attending conference ISBRA 2015.

• APRIL 2013: Ph.D. fellowship from Tata Consultancy Services (TCS) 

• JULY 2012: Ph.D. fellowship from Indian Institute of Technology, Kharagpur (until March 2013).