Computational Biology and Bioinformatics

Build a federated sepsis early-warning model, test privacy budgets, and measure how data splitting changes prediction quality and utility.

Reanalyze public single-cell RNA-seq data to find rare cell states with contrastive learning, marker genes, and validation in GEO datasets.

Build and score generative PROTAC linker designs for KRAS-G12D, then test how modeling choices change ternary-complex confidence and ranking.

Compare mammal genomes to rank DNA-repair gene losses, build phylogenetic thinking, and practice statistical filtering for longevity research.

Build a connectome simulation of C. elegans, test perturbations, and learn how a minimal circuit can reproduce chemotaxis.

Analyze spike gene evolution with phylogenetics and dN/dS tests to find animal reservoir mutations that may raise spillback risk.

Model measles spread with county exemption data, find vaccination tipping points, and build simulation, clustering, and sensitivity analysis skills.

Model how plant circadian clocks respond to changing day length, and use public phenotype data to predict which Arabidopsis ecotypes perform best.

Build a contrastive model that links SMILES and bioassay text, then test how well it retrieves drug candidates from PubChem-style data.

Reanalyze ribosome profiling data, predict micropeptide structures, and rank smORFs in lncRNAs with a bioinformatics project on gene discovery.