You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I used cellSNP to identify SNP of scRNAseq. But I found a problem many SNP sites are not in gene. In my opinion, All Reads of scRNAseq should come from gene, So all SNP sites should locate in gene.
The text was updated successfully, but these errors were encountered:
Good quesiton. It actually depends on the protocol of the sequencing, and many of them are far from the expected and return many reads mapped to integenic regions.
Also, I assume that the genome build is matched between the VCF and bam file.
I used cellSNP to identify SNP of scRNAseq. But I found a problem many SNP sites are not in gene. In my opinion, All Reads of scRNAseq should come from gene, So all SNP sites should locate in gene.
The text was updated successfully, but these errors were encountered: