Dabin Jeong
AI for Life Science Researcher
Wellcome Sanger Institute, Cambridge, UK- ✉️dj16@sanger.ac.uk
GitHub
LinkedIn
Google Scholar
About me
I am an AI for Life Science researcher at the Lotfollahi Lab, Wellcome Sanger Institute, working at the intersection of machine learning and biomedical science.
I see myself as a problem finder as much as a problem solver. My background in bioinformatics and biochemistry leads me to look for gaps where existing computational approaches fall short. Rather than simply applying established AI frameworks to biology, I focus on identifying under-explored problems in the domain and developing novel solutions to address the problems. I translate the complexity of biological and medical questions into well-defined computational problems, then build the models needed to solve them.
💡 Research Interest
- Multi-modal learning of biomedical images and spatial transcriptomics
- Regulatory DNA sequence generation
- Temporal cell dynamics modeling
🚩 Research Vision
“What I cannot create, I do not understand” — Richard Feynman
I was drawn to computational biology by the idea that DNA is the ultimate code: a sequence in which both code and the codebook are deeply entangled. My research ambition is to build a program that can simulate biological systems from DNA sequence.
🔥 News
- 🎉 Sep 2026 · LocoGen, a locality-aware generative model for designing regulatory DNA, was accepted to NeurIPS 2026.
- 🎉 Jul 2026 · Selected as one of 500 participants from over 6,000 applicants in the Anthropic × Gladstone Claude Life Sciences Hackathon, where I built a generative model that infers missing cell states across time points (LinkedIn post).
- 🎉 Jun 2026 · SIGMMA was accepted to the ICML FM4LS/SD4H workshop; a journal version coming soon!
- 🎉 Oct 2024 · Joined the Wellcome Sanger Institute.
💼 Research & Industry Experience
Senior Data Scientist · Wellcome Sanger Institute (Lotfollahi Lab), UK Oct 2024 – present
SIGMMA
Cross-modal method aligning H&E histopathology with spatial transcriptomics.
LocoGen
Property-guided generative framework for de novo regulatory DNA design in a lab-in-the-loop cycle.
Research Intern · LG AI Research (Multimodal Lab), Korea Jan – Apr 2024
EXAONEPath
Self-supervised foundation model for H&E histopathology images.
TCGA multi-omics pipeline
Scalable, reproducible preprocessing pipeline for TCGA multi-omics data.
Graduate Researcher · Seoul National University (Biohealth Informatics Lab), Korea Sep 2021 – Aug 2024
GOAT
Graph attention network for gene-level biomarker discovery in eosinophilic asthma subtypes.
VWA5A as a biomarker in breast cancer
Machine-learning-based protein prioritization identifying a metastasis biomarker.
Condition-specific GRN
Kernel canonical correlation analysis to simulate transcriptomic responses to external stimuli.
Patent (Korea)
Multi-omics biomarker discovery method, registered through KIPRIS.
🎓 Education
Ph.D., Interdisciplinary Program of Bioinformatics · Seoul National University, Korea · Sep 2018 – Aug 2024
- Thesis: Computational modeling of molecular interactions for biomarker discovery using single/multi-omics data
- 3 first-authored and 10 co-authored papers
B.S., Biochemistry · Yonsei University, Korea · Mar 2013 – Aug 2018
Technical Skills
- Programming · Python, R, Shell, LaTeX
- ML / Deep Learning · PyTorch, PyTorch Geometric, DGL, scikit-learn
- Workflow & Reproducibility · Nextflow, Snakemake, Docker, Git
Awards & Achievements
- 🏆 Invited talk (2023) · National Heart & Lung Institute, Imperial College London (host: Prof. Kian Fan Chung) — Deep graph attention for multi-omics network biomarkers in eosinophilic asthma
- 🏆 National Excellence Scholarship (2016) · Natural Sciences and Engineering; sole recipient in the department
📷 Photo
