Genome-Wide SNP Disease-Risk Prediction

An elastic-net logistic regression and a random forest estimate, from a person's SNP genotypes, the risk of five common diseases: type 2 diabetes, Alzheimer disease, type 1 diabetes, asthma and atrial fibrillation. The risk SNPs and their odds ratios come from the GWAS Catalog, and the genotypes are real 1000 Genomes haplotypes. Each model is compared with a classical polygenic risk score and combined with it in a stack. The disease labels are simulated, because the GWAS Catalog holds only summary statistics. A complete stand-alone MATLAB R2026a version is included.

Category: Bioinformatics - Genomics and Sequence Analysis (Machine Learning, Statistical Genetics, MATLAB, Python)

Bioinformatics Project B10 from ProTechSC Academy, with graphical abstract, video demonstration and project guidance.