Is your ranking biology, or arithmetic?

Drop the table your screen produced. This measures how much of its order is explained by how the gene sets were built — chiefly their size — with no reference to what any gene does.

Drop your results file here or click to choose · CSV, TSV or TXT
MAGeCK · BAGEL2 · drugZ · g:Profiler · DAVID · clusterProfiler · Enrichr · fgsea · GSEA

Nothing to hand? — the 50 Hallmark programs from the study behind this tool. It grades us at 46%, which is worse than nine in ten published screens.

Your file never leaves this browser. There is no server and no upload: the Python that runs the check is downloaded to your machine and runs there. You can disconnect after the page loads and it still works.

The verdict

Against the field

What moves when the correction is applied

Look up a published screen's floor

The atlas carries the no-biology floor for 1,272 published human CRISPR screens — the share of each one's gene-set ranking that set size alone predicts. Enter a BioGRID ORCS screen id to get the number, the method, and the string to cite. Looked up, never recomputed, so your copy and everyone else's agree.