Computational biologist developing machine learning methods for translational genomics and precision oncology.
My work focuses on extracting clinically meaningful regulatory and phenotypic signals from complex biological data, with a current emphasis on cell-free DNA, liquid biopsy, and multi-omic inference. I am especially interested in methods that are mechanistically grounded, uncertainty-aware, and useful in real-world translational settings.
- Regulatory-state inference from minimally invasive assays
- Machine learning for tumor phenotyping, expression inference, and treatment-response monitoring
- Multi-omic modeling with an emphasis on rigor, interpretability, and translational relevance
- Open and reproducible research software for computational oncology
Fragmentomic and phased-nucleosome analysis of cfDNA from BAM/CRAM data, with region-level feature extraction and signal profiling for downstream biomarker discovery and modeling.
Cancer subtype classification and heterogeneous phenotype fraction estimation from cfDNA-derived features using reference models built from biologically anchored training data.
Deep learning framework for predicting tumor gene-expression and regulatory programs from cfDNA-derived signal profiles, with emphasis on interpretability, robustness, and uncertainty calibration.
-
Deep learning-based non-invasive profiling of tumor transcriptomes from cell-free DNA for precision oncology
Preprint / manuscript in review -
Nucleosome patterns in circulating tumor DNA reveal transcriptional regulation of advanced prostate cancer phenotypes
Cancer Discovery (2023) -
A framework for clinical cancer subtyping from nucleosome profiling of cell-free DNA
Nature Communications (2022) -
Preeclampsia risk prediction from prenatal cell-free DNA screening
Nature Medicine (2025)
I aim to build computational methods that move beyond static biomarker detection toward dynamic, mechanistically informed inference of tumor and tissue state. Long term, I am interested in scalable models that connect genomics, epigenomics, and transcriptomics in ways that are both scientifically rigorous and clinically actionable, with the power to reveal new insights around fundamental biology.