Read is confidently mapped to the transcriptome in Cell Ranger

Question

[tien]

Dear all,

I'm trying to read and understand tags from BAM file resulted by Cell Ranger.

In section "Transcriptome alignment", they consider all reads that map to exon or intron or intergenic region. https://support.10xgenomics.com/single-cell-gene-expression/software/pipelines/latest/algorithms/overview#alignment

And they define tag xf=1 for reads that confidently map to transcriptome.

I'm wondering if:

• Reads that map to intronic or intergenic region can have xf=1
• Reads that map to intronic or intergenic region can have value for tag GN (gene name) and GX (gene id)

Since I cannot find any reads that satisfy any of above two conditions. If it's true to have no such read, then does it violate what they talk about "transcriptome alignment"?

Moreover, I observe, from my data , that reads have MAPQ < 255 (low quality mapping) will not have xf = 1. Is it always true in general?

Thanks for your help.

[GenoMax]

Reads that map to intronic or intergenic region

This may be dependent on version of cellranger you used. It is only with v.7.0 they started including intronic reads by default. See: https://support.10xgenomics.com/docs/intron-mode-rec

[tien]

Hi, thanks for your reply. I think the link is about counting (expression). However I'm talking about alignment result (BAM file).

[GenoMax]

Correct but unless you used the latest version intronic reads would not be included at all.

Answer 1

[tien]

I think I found the answer

• Annotation GTF have no information for intronic region or intergeneic region, so we cannot have GN or GX for those reads
• We need to set --include-intron to have those intronic reads aligned to transcriptome