Deep Learning · PLMs
Compat Net: Interaction aware protein compatibility prediction model
A multi-task deep learning framework using protein language models and transformer-based gated cross-attention to model viral–host protein interactions.
Computational biology · Machine Learning. Artificial Intelligence
My research focuses on developing machine learning methods that learn from large-scale biological data and limited experimental observations to model protein interactions, viral evolution, and antigenic escape, with the goal of improving infectious disease prediction, surveillance, and intervention.
Mizzou · CS
01 — About
I'm a Ph.D. student in Computer Science at the University of Missouri-Columbia. My work sits at the intersection of artificial intelligence and computational biology, where I develop AI systems that learn from complex, sparse, and large-scale biological data.
My research focuses on protein interaction prediction, viral evolution, antigenic escape, vaccine effectiveness, and infectious disease modeling. I use deep learning, protein language models, foundational models, and scalable bioinformatics pipelines to transform biological data into predictive insight for infectious disease surveillance and intervention.
02 — Research
A selection of research projects that showcase my work at the intersection of AI and computational biology.
Deep Learning · PLMs
A multi-task deep learning framework using protein language models and transformer-based gated cross-attention to model viral–host protein interactions.
Graph Neural Networks
Graph neural network models that capture epistatic mutation effects underlying influenza antigenic evolution and immune escape.
Genomics · Vaccines
Analysis of viral sequence diversity, polymorphisms, entropy, and antigenic evolution to support improved influenza vaccine design.
03 — Publications
Prediction models for COVID-19 disease outcomes.
Intrahost HA polymorphisms and culture adaptation shape antigenic profiles of H3N2 influenza viruses.
Epitope-spanning antigenic variation reprograms immunodominance and broadens immunity in sequential influenza vaccination.
Species- and variant-specific ACE2 compatibility shapes SARS-CoV-2 spillover potential in North American cervids.
Interaction-aware multitask deep learning reveals cross-species receptor compatibility landscapes across sarbecoviruses.
04 — Path
Professional Experience
Aug 2022 – Present
University of Missouri-Columbia
Developing AI and bioinformatics methods for protein binding prediction, antigenic escape modeling, viral genomic analysis, vaccine effectiveness, and differential gene expression analysis.
Feb 2020 – May 2022
PSP Corporation — South Asia Liaison Office, Nepal
Built deep learning models for medical image segmentation: multi-organ segmentation, pseudo-labeling for cost-efficient annotation, and lung fissure detection from real patient imaging.
Education
Aug 2022 – Present
University of Missouri-Columbia
2015 – 2020
Tribhuvan University, Nepal
05 — Service
University of Missouri · 2025–2026
Led graduate student engagement initiatives, including the Mizzou Tiger Internship Seminar, ShowMe Course Project Showcase, and a college-wide ARC-AGI hackathon.
University of Missouri · 2024–2025
Missouri University Nepali Student Association · 2025–2026
University of Missouri · 2025–2026
Represented the EECS department and contributed to student affairs initiatives supporting graduate student mental health and well-being.
06 — Recognition
2026
EECS Department, University of Missouri
April 2026
2024
Graduate Professional Council, University of Missouri
2024
2026
Santa Fe Institute
Selected Research Talk
07 — Talks
Epidemics Conference 2025
Intrahost polymorphisms and antigenic adaptation in H3N2 influenza viruses.
American Society of Virology (ASV 2024)
Differential expression analysis of innate immune responses.
08 — News
Add your own moments here — talks, papers, travel, lab events.
2026
Recognized by the EECS department at the University of Missouri.
2026
Selected for a research talk at the Santa Fe Institute.
2025
Presented a poster on H3N2 intrahost polymorphisms and antigenic adaptation.
09 — Contact
Open to collaborations, talks, and conversations about computational biology, viral evolution, and machine learning for the life sciences.