You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
There are four prevalence filters we can apply to the feature table.
189
-
Two of these, `--p-min-labeled-sources` and `--p-min-labeled-sources` filter features based on their prevalence across the retained sources.
189
+
Two of these, `--p-min-unlabeled-sources` and `--p-min-labeled-sources` filter features based on their prevalence across the retained sources.
190
190
The interpretation of the values provided in the above command is that if a feature is present in less than 6 unlabeled sources or less than 3 labeled sources, then it will be filtered.
191
191
Because we decided to retain 6 and 3 unlabeled and labeled sources respectively, we are requiring retained features to present in *all* samples.
192
192
The other two prevalence filters, `--p-min-unlabeled-fractions` and `--p-min-labeled-fractions` filter features based on fraction (sample) prevalence.
0 commit comments