CS · AI · Computational Biology
From protein mutations to tumor immune microenvironments to ecological systems — I build AI instruments grounded in the AIVC framework that reason across the scales of biological complexity.
Uncertainty-aware decoder virtual instrument for immunotherapy response prediction in melanoma. scVI + GAT + MC Dropout.
GNN-based protein mutation effect prediction. Single-point mutations, protein stability (ΔΔG) and antibody-antigen binding affinity.
AI research assistant for the AIVC literature. Natural language queries across papers, cross-paper synthesis via knowledge graph.
A maker build log — sensor stack, Raspberry Pi edge compute, and a bridge to AIVC concepts. Treating ecology as a biological system.
● In Build Comp. HerbologyField collection → LC-MS → LINCS L1000 → Virtual Cell perturbation prediction. Bridging traditional herbal knowledge and modern computational biology.
● Research