Discrepancy in TSS enrichment score between cellranger and Signac: am I using the wrong annotation? #1682
Replies: 1 comment 1 reply
-
Answering to myself after some research. Devs please correct me if I am again wrong, otherwise the topic can be closed. I was using the wrong annotation, the EnsDb.Mmusculus.v79 is based on the GRCm38 (you get this info if you just run EnsDb.Mmusculus.v79 as a command in R after loading the library) and the reference for my dataset was GRCm39. For anyone running into a similar issue in the future, this is what I did to solve the problem: I used AnnotationHub package to query the Annotation Hub for the available annotations for my genome and release:
I chose the ID of the appropriate GTF file from the query output and built an Ensembl Annotation package using functions from the ensembldb library:
Finally, I prepared the annotations object as in the Signac vignettes:
Cheers, |
Beta Was this translation helpful? Give feedback.
-
Hi!
First of all thank you for the great documentation and vignettes!
I have a scATACseq project for which I got the cellranger output including the HTML summary. In that summary, GRCm39.e104 is stated as reference. The cellranger output of the TSS enrichment score reports it as 10.4, while the mean of the TSS enrichment score as calculated by
TSSEnrichment()
is at 1.4 (median is also 1.4). I am using the workflow as in the vignette for human PBMCs, adapted to mouse (EnsDb.Mmusculus.v79 as reference annotation). Am I missing something? Should I be using a different reference annotation?Thanks a lot in advance!
Beta Was this translation helpful? Give feedback.
All reactions