Differential analysis & tests / Multiple-testing correction
Your data stays in your browser: files are read and processed on your device, never uploaded to any server, and cleared when you close the page. Free, no sign-in.
When many tests are run at once (e.g. thousands of genes), calling P < 0.05 significant yields many false positives, so P values need adjusting.
Benjamini–Hochberg controls the false discovery rate (FDR) and is the usual choice in omics; Benjamini–Yekutieli stays valid under any dependence but is more conservative; Bonferroni, Holm and Hochberg control the family-wise error rate (probability of any false positive) and are stricter.
Rows with missing P values stay in the output with missing adjusted values and are not counted in m (as R p.adjust by default) — appropriate when a missing value means “not tested”.
The histogram shows raw and adjusted P values: a spike near 0 over a flat background indicates real signal; a pile-up towards 1 suggests the test assumptions may not hold.
Benjamini Y. & Hochberg Y. (1995) J R Stat Soc B 57:289–300; Benjamini Y. & Yekutieli D. (2001) Ann Stat 29:1165–1188; Holm S. (1979) Scand J Stat 6:65–70; Hochberg Y. (1988) Biometrika 75:800–802. Results match R p.adjust and statsmodels multipletests.
Up to 500,000 rows.
Send us your data and research question and you will receive a written plan within 1 working day: analysis steps, parameter rationale, deliverables and timeline. Quoted per project.