Reads in IGV are mostly in intron regions

Question

[Kai_Qi]

I was analyzing a published datasets related to Patch-seq: RNAs from soma or dendrites was sequenced.

After visualizing the BAM files in IGV, it is weird to see that most of the reads fall into intron regions. I am wondering if this could be the problem of the data quality considering the small amount of RNA?

Thanks,

[Pierre Lindenbaum]

check the genome build/reference you're using...

[Ming Tommy Tang]

ha, this happened to me. check the reference genome that you used to align and make sure you use the same version in IGV.

[Kai_Qi]

Thanks for the advice. I have ben used the GRCm38_P4 and GRCm38_P10 to build the reference and to do the mapping, and in IGV I used mm10 for the visualizing. It looks both reference give me the same situation. I am wondering if this could be genome contamination.

[Ming Tommy Tang]

do you see it for all genes? do you see any exonic mapping?

[Kai_Qi]

For high abundance genes such as Actb and ribosome gene Rps29 it looks fine: Actb the reads fall into 3'UTR while the Rps29 the read is the last 2 exons (3 exon in total of this gene). Nevertheless, some classical neuronal genes which supposed to be high in the neurons captured doesn't look good. Thank you for your replies.

[Ming Tommy Tang]

you should run RNAseqQC tools to get the percentage of exon reads. If it is too low, something maybe wrong with your experiment.

[Kai_Qi]

Thanks you for this suggestion. I will try to run the RNA-seQC and see how it turns to be

[Ming Tommy Tang]

This one should be good https://pubmed.ncbi.nlm.nih.gov/33677499/