Hello,
I was wondering if someone knows the differences between the counts and data parts of a Seurat object.
Hello,
I was wondering if someone knows the differences between the counts and data parts of a Seurat object.
When you first load up the data both counts and data contain the raw UMI counts. If you normalize the UMI counts using NormalizeData the results will be stored in data.
Normalization via SCTransform is similar in that the raw UMI counts are stored in counts, and the corrected counts in data but of the SCT assay instead of the (default) RNA assay. Note that SCTransform corrected counts are not recommended to be used for anything but dimension reduction and clustering, although v2 might change this recommendation soon [SOURCE].
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.