Bioinformatics graduate with research experience in computational genomics, sequencing-data analysis, RNA biology, and machine learning. I build scalable HPC workflows for WGS, fragmentomics, epigenomics, and AI-driven structure prediction using Python, C++, Bash, and deep-learning frameworks.
Coursework: Bioinformatics I & II, Genomics, Machine Learning, Deep Learning, Algorithms for BCB, Biochemical Pharmacology.
Coursework: Python, C++, Embedded Systems, AI Fundamentals, Circuit Python, Signal Processing, Diagnostic Instrumentation.
Projects span cancer genomics, biomarker discovery, RNA structure prediction, tool development, sequencing pipelines, multi-omics analysis, and biological machine learning.
Built a VFold-style RNA motif database and toolset: motif extraction, sequence grouping, RMSD-based structural clustering, representative-template selection, and a template-search tool for RNA 3D modeling.
Template search: R1116 decomposed into 17 motifs, each matched to PDB templates (1txs, 1c2w, 6mtb, ...)
Modified OpenFold3 (9 source files) and Boltz-2 (2 source files) to accept RNA templates (capabilities the stock models lack), and built a secondary-structure-driven synthetic-MSA generator. Benchmarked all approaches against unmodified baselines by C1' RMSD.
1E7K: baseline 8.98 Å vs synthetic MSA 0.94 Å (overlay: reference black, prediction green)
Built a two-stage FASTQ-to-BAM WGS pipeline (BWA + SAMtools workers on SLURM/LSF), with fragment-length extraction, 4-mer/5-mer end-motif analysis, multi-tier QC, and cancer-vs-healthy cohort comparison across Streck/EDTA tubes and WGS/HMC assays.
WGS short/long fragment ratio: cancer cohort shifted higher than healthy
WashU (Maher Lab): Fragmentomics & cfDNA: Built FASTQ-to-BAM WGS pipelines on LSF, fragment-length and 4/5-mer end-motif analysis, multi-tier QC, and cancer-vs-healthy cohort comparison across Streck/EDTA tubes and WGS/HMC assays.
SLU (Hou Lab): RNA structure prediction: Modified OpenFold3 and Boltz-2 at the source level for RNA template support, built a synthetic-MSA generator, and benchmarked predictions on a 22-RNA set, reaching sub-Ångström accuracy on favorable cases.
ML project: ESM-2 embeddings + engineered biological features in an ensemble model (ROC-AUC ≈ 0.80), with SHAP-based interpretation of feature contributions.
IEEE ASIANCON 2023: IoT-enabled smart stethoscope with biomedical signal acquisition, filtering/amplification circuitry, and Arduino-based instrumentation.
I am interested in roles involving sequencing-data analysis, computational genomics, RNA biology, cancer genomics, biomarker analysis, bioinformatics tool development, and machine learning for biological systems.