How are intron-containing transcripts captured in snRNA-seq?

Question

[pomodoro_sinensis]

Hello everyone,

I am a bit confused about how one can get reads coming from introns in snRNA-seq. Usually, transcripts are captured through the poly(A) tail, which only gets added to transcripts once transcription is over. However, splicing occurs co-transcriptionally, so in theory, no intron should be left once poly-adenylation has occurred.

I understand that this might not be the case for all transcripts and conditions, and that splicing in some transcripts can occur slow enough to allow this poly-adenylated intermediates to be captured. However, wouldn't this mean that this method is biased towards those transcripts? Is there any study that explores this, or has this ever been discussed somewhere?

Thank you!!

Answer 1

[rpolicastro]

Instead of the oligo(dT) priming from the poly(A) tail, it accidentally primes from genomic poly(A) runs found in introns, thus accidentally capturing nascent RNA. RNA-velocity methods such as Velocyto[1] and scVelo[2] use this information to infer transcription kinetics and infer directional pseudotime. Note that RNA-velocity is still a developing analysis[3] and should be used with caution[4].