Research Areas
- Computational cancer systems biology
- Multi-modal graph representation models for precision oncology
- Dynamic graph neural networks for Alzheimer’s disease progression
Scientific Achievements
- Developed an integrative graph neural networks (IGCN) framework leveraging multi-omics data to enhance precision oncology and identify novel therapeutic targets.
- Designed advanced deep learning architectures for early prediction of Alzheimer’s disease progression using longitudinal EHR, facilitating timely intervention and improving patient care outcomes.
- Chosen to serve as a mentor for the NIH All of Us Research Program, 2024.
- Accepted as an ACCEL Scholar for the NIGMS-funded DE-Clinical and Translational Research program.
Funding
Fundings obtained with RCMI support:
- Appointed as a Senior Data Scientist through RCMI award U54MD015959
- Accepted as an ACCEL scholar under DE-CTR ACCEL program U54GM104941
Scientific Advance
IGCN: Integrative Graph Convolution Network for patient level insights and biomarker discovery
Published in Oxford Bioinformatics , Volume 41, June 2025, PMCID: PMC112104196.
Published in Oxford Bioinformatics , Volume 41, June 2025, PMCID: PMC112104196.
This study introduces the Integrative Graph Convolutional Network (IGCN), a novel computational framework for multi-omics analysis in cancer. By integrating diverse molecular networks, IGCN captures complex interactions across genomic, transcriptomic, and proteomic layers, enabling interpretable and highly accurate predictions of cancer molecular subtypes. Benchmarking against existing methods demonstrated superior performance, while the model also uncovered molecular level biomarkers, offering actionable insights for precision oncology and advancing personalized medicine.
