When should I correct for batch effects
1
2
Entering edit mode
NeuroDaniel ▴ 20
@9c42215f
Last seen 4 months ago
Canada

Hello,

I am working on snRNAseq data but I am very new to it, so I am still trying to figure some things out. I followed the OSCA book for my analyses but I am still unsure as to when to apply batch corrections.

I have four experimental groups (three mouse brains pooled per group), but I understand correctly, during the actual snRNAseq experiment (done by a collaborator) there were four Illumina libraries prepped (one per group, same protocol for all done at the same time) and then they were all sequenced together in a NovaSeq 6000 S4 flow cell. Then, they were demultiplexed using CellRanger and thus I got four filtered matrices.

What I am still unsure about is whether I need to apply batch corrections because the libraries were sequenced together. I applied MNN batch correction but realized that whether I applied it or not, my clustering and tSNE projections were virtually the same.

Do I need to apply batch correction in this case? I'm sorry this is not really a coding question, rather a technical one, but I appreciate the help!

Thank you!

BatchEffect SingleCellExperiment • 502 views
ADD COMMENT
2
Entering edit mode
@james-w-macdonald-5106
Last seen 8 hours ago
United States

You have pretty much already answered your own question. The way I would normally check is to cluster separately, identify the clusters (for brain it should be easy, using SingleR and the relevant existing dataset), and then combine and see if the cell types cluster together after you combine. If they do, you should be good. If not, you need to batch correct.

Login before adding your answer.

Traffic: 381 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6