Home▸Molecular Biology▸Multi-Omics Patient Subtyping — 2D Joint NMF

Multi-Omics Patient Subtyping — 2D Joint NMF

A 2D multi-panel view of joint non-negative matrix factorization fusing synthetic genomics, transcriptomics and proteomics data into a shared latent space, with a draggable/zoomable factor-space scatter, a sorted patient x feature heatmap, and a live convergence chart.

Molecular Biology2DAdvanced60 FPS📱 Mobile-adapted⇄ 3D version

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. This 2D layout lays the same computation out across three linked panels: a draggable, zoomable factor-space scatter you choose the axis pair for, a cluster-sorted patient × feature heatmap that reveals the block structure being fitted, and a live convergence chart tracking reconstruction error and silhouette score — the same computational pattern behind real multi-omics cancer-subtyping pipelines.

⚙ Under the hood

A 2D multi-panel view of joint non-negative matrix factorization fusing synthetic genomics, transcriptomics and proteomics data into a shared latent space, with a draggable/zoomable factor-space scatter, a cluster-sorted patient x feature heatmap, and a live convergence chart tracking reconstruction error and silhouette score.

multi-omicsmatrix factorizationclusteringbioinformaticssystems biologydisease subtyping

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install