Filtering nuclear RNA expression instead of MT expression in Scanpy

Question

[wonge29]

I was wondering if anyone knows if there is a straightforward way to filter for nRNA expression. I know Scanpy is able to filter mitochondrial reads by first identifying using ["MT"]. Is there a way to do so for nuclear RNA? Or would I need to download anything.. I want to retain MT for certain reasons rather than filtering out high expressions of MT.

[Ram]

You asked this question under 3 hours ago on Bioinfo SE. Any reason you're cross-posting it here so soon? Cross-posting tends to annoy both communities involved.

Answer 1

[ATpoint]

The MT (prefix) method is actually suboptimal as it will fail once you are not using the trivial gene names but something like Ensembl gene IDs (ENSMUSG...). I personally prefer to load the GTF reference file and then scan for genes that sit on the mitochondrial chromosome (or any other kind of gene type I want to filter for).

As for your problem: What exactly do you mean by nRNA? Is it non-mitochondrial RNAs, or small nuclear RNAs? Please elaborate. If it is anything but the mitochondrial RNA then load a GTF file and select all genes not located on chrM.

Does that make sense?

[mt_pereira]

Hello! I am having the problem that you have mentioned exactly because my mitochondrial genes do not have the MT prefix.

I opened an issue here https://stackoverflow.com/questions/75172257/is-there-a-command-to-annotate-genes-in-a-matrix-from-a-txt-file?noredirect=1#comment132652504_75172257

Would be very greatful if you could help!