Computational Biology and Bioinformatics

Build causal graphs from wearable and diet logs to test how sleep, glycemia, and mood may influence one another in teens.

Build and test a protein-complex modeling pipeline using AlphaFold-Multimer and cross-link mass-spec data, then score accuracy against PDB.

Compare asthma polygenic risk scores across ancestries, test transfer-learning corrections, and measure whether AUC improves in underrepresented groups.

Build a network pharmacology project that maps herbal compounds, targets, and diabetes pathways, then tests synergy hypotheses with public data.

Build a voice-classification project that uses contrastive learning to find early prosodic patterns linked to Parkinson’s and Alzheimer’s.

Build an agent-based model that links prescribing norms to AMR allele shifts, then test how behavior spread changes resistance patterns.

Use public sequences and city location data to test how urban heat islands may shape genetic patterns in common animals.

Build an NLP pipeline that finds early disease signals in public posts, then test how well it predicts known outbreak events.

Model yeast metabolism with COBRApy, then test how ethanol, salt, and caffeine change fermentation rate using simple CO₂ measurements.

Build a smartphone image pipeline that classifies pond microbes, estimates biodiversity, and turns simple microscopy into real-time water-quality data.