From 8fa2f2f832ea24f980cd29d30da062de6c1c850f Mon Sep 17 00:00:00 2001 From: Michaela Mueller Date: Fri, 11 Apr 2025 13:41:13 +0200 Subject: [PATCH] add short explanation on LISI scaling --- src/metrics/lisi/config.vsh.yaml | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) diff --git a/src/metrics/lisi/config.vsh.yaml b/src/metrics/lisi/config.vsh.yaml index 7b852431..34ce17ca 100644 --- a/src/metrics/lisi/config.vsh.yaml +++ b/src/metrics/lisi/config.vsh.yaml @@ -14,7 +14,8 @@ info: variable. The original LISI score ranges from 0 to the number of categories, with the latter - indicating good cell mixing. This is rescaled to a score between 0 and 1. + indicating good cell mixing. This is rescaled to a score between 0 and 1, where 0 + represents bad batch mixing and 1 represents good batch mixing. references: doi: 10.1038/s41592-021-01336-8 links: @@ -33,7 +34,10 @@ info: distance computation on single-cell kNN graphs. The metric assesses whether clusters of cells in a single-cell RNA-seq dataset are well-mixed across a categorical cell type variable. - The original LISI score ranges from 0 to the number of categories, with the latter indicating good cell mixing. This is rescaled to a score between 0 and 1. + The original LISI score ranges from 0 to the number of categories, with the latter + indicating good cell mixing. This is inverted and rescaled to a score between 0 and 1, + where 0 represents bad cell type separation and 1 represents good cell type separation. + references: doi: 10.1038/s41592-021-01336-8 links: