This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Can I estimate RNA velocity in Visium Spatial Gene Expression for FFPE

Dear USERs,

I have following question. Can I estimate RNA velocity in Visium Spatial Gene Expression for FFPE?

To estimate RNA velocity, we need to calculate the unspliced & spliced RNAs in a gene. I did it using velocyto program. But, they couldn't do it.

So, can we obtain the unspliced/spliced RNA correctly from Visium for FFPE (v2)?

https://cdn.10xgenomics.com/image/upload/v1666737555/support-documents/CG000605_GenomicDNAinVisiumCytAssistSpatialAssay_RevA.pdf

visium rna

1 answer

No, Visium uses probes that exclusively (to my knowledge) binds exons, so no unspliced counts from that assay.

A fun experiment might be to see how many unspliced counts you get (probably not many at all -- but still would be cool to check since exons are also part of unspliced RNA). Edit: Maybe I'll try this analysis myself in the next week or two since I'm working on a similar project.

Thanks! Then, did you check it?? I'll try it as soon as possible.

Yes! Here, are the results I have:

Nascent RNA (i.e. intron-containing reads): 0.49% (i.e. less than 1 percent).

This was from 10X's CytAssist_11mm_FFPE_Mouse_Embryo dataset.

For comparison, on single-cell RNA-seq reads, I usually get 25% nascent RNA; on single-nucleus RNA-seq reads, even higher.

It is so interesting that they have small portion of nascent rna. However, I don't think we can use Visium data to estimate velocity (it has too small proportion). What do you think about that?

Actually, I did it in last year, I failed estimating velocity using velocyto. From your response, I guess the small proportion is the reason why I failed.

Definitely not -- you can't use visium to estimate velocity, since there's hardly any intronic content.

+1

The reason, @oghzzang as said above, is that it uses probes that binds to predefined exonic regions. Unlike the "normal" single-cell sequencing with polyA enrichment where the primers have lots of random priming to introns, probes do not.

Yes, I absolutely agree with you. Thank you!

Yes, your comment helped me a lot! thank you!

Log in to answer this question.