The world's lab benches, food matrices, biospecimens and literature — distilled into thousands of citation-backed media, every component mapped to a standard BiGG exchange reaction, so any genome-scale model can adopt a medium in one line.
Every medium is already resolved to standard BiGG exchange reactions with bounds — no manual mapping, no unit wrangling. Pull it straight into COBRApy and constrain your model.
Search and filter every medium by category, source, food group and oxygen. Click any one to inspect its full component list, cross-references and citation — then copy it into COBRApy.
Browse all media →Put two to eight media side by side. See a Jaccard similarity matrix, an UpSet intersection plot, and a clustered grid of exactly which compounds they share and where they diverge.
Compare formulations →A hierarchically clustered view of the whole collection — which compounds co-occur, and how media groups separate by composition. Row- and column-clustered heatmaps with dendrograms.
See the clustering →The distribution of coverage across the whole library — the share of each medium's compounds that resolve to a standard BiGG exchange. The library is near-complete: almost every medium sits in the high-confidence band. The spectrum bar below splits the collection by reliability. Hover a bar for its count.
Sorted by curation tier — our ★ curated reference collection leads the frontier, then expert-curated ★, paper-verified ✓, database-sourced ●, and auto-extracted ⚠. Every medium is cited and mapped through the same pipeline. Click a row to open its full detail view.
Data & schema: provenance rules · programmatic API · every medium retains its original citation.