Multi-Omics Patient Subtyping via Joint Matrix Factorization
Watch a joint non-negative matrix factorization decompose synthetic genomics, transcriptomics and proteomics data into shared latent factors, then k-means clusters patients into disease subtypes in 3D factor space with live reconstruction error, silhouette score and subtype purity.
A cohort of synthetic patients carries three simulated omics layers — genomics, transcriptomics and proteomics — generated from a shared hidden biology plus realistic per-layer batch noise. A live joint non-negative matrix factorization (Lee–Seung multiplicative updates) fuses all three layers into one shared latent space, and k-means clustering on that space groups patients into candidate disease subtypes, rendered as colored points in 3D. Live readouts track reconstruction error, silhouette score and subtype purity against the hidden ground truth as the factorization converges — the same computational pattern behind real multi-omics cancer-subtyping pipelines.
Watch a joint non-negative matrix factorization fuse synthetic genomics, transcriptomics and proteomics layers into a shared latent space, then k-means clusters patients into disease subtypes in 3D with live reconstruction error, silhouette score and subtype purity.
3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install