From 6f3f398b7898ddc75cec9214a0b5764b9094127e Mon Sep 17 00:00:00 2001 From: Lukas Heumos Date: Tue, 4 Aug 2026 17:15:22 +0200 Subject: [PATCH] Rename DistanceTest alpha to padj_threshold --- distance_tests.ipynb | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/distance_tests.ipynb b/distance_tests.ipynb index a138ab7..8bd353e 100644 --- a/distance_tests.ipynb +++ b/distance_tests.ipynb @@ -156,7 +156,7 @@ } ], "source": [ - "etest = pt.tl.DistanceTest(\"edistance\", n_perms=1000, obsm_key=\"X_pca\", alpha=0.0015)\n", + "etest = pt.tl.DistanceTest(\"edistance\", n_perms=1000, obsm_key=\"X_pca\", padj_threshold=0.0015)\n", "tab = etest(adata, groupby=obs_key, contrast=contrast)" ] }, @@ -165,7 +165,7 @@ "metadata": {}, "source": [ "We recommend using the adjusted p-values, which are corrected for multiple testing.\n", - "You might have to specify a lower `alpha` value for the adjusted p-values to be significant, depending on the number of permutations, the number of cells in each group, and the number of groups tested (each group increases the number of tests made and therefore affects the adjusted p-value).\n", + "You might have to specify a lower `padj_threshold` value for the adjusted p-values to be significant, depending on the number of permutations, the number of cells in each group, and the number of groups tested (each group increases the number of tests made and therefore affects the adjusted p-value).\n", "Also note that the lowest p-value you can get is 1 / n_permutations, which is the probability of getting the observed distance by chance." ] },