Strand specificity

[flde]

Hello,

many thanks for the great tool!

I am using 10x 5' with R2-only mapped reads - hat should be antisense mapped reads only. And cellSNP-lite in mode 1a with 1k genome vcf. I assume the 1K genome vcf SNP reference to the positive-sense strand.

My question is if cellSNP does take into account the orrientation of the reads relative to reference SNP? I assume yes but highly appreciate your insights.

Kind regards,
Florian

[hxj5]

Hi, cellsnp-lite does not check the orientation of the reads. It extracts the SNP allele from the SEQ field of the BAM alignments, which should always be represented on the forward genomic strand (i.e. the same strand as the reference genome sequence). In this way, cellsnp-lite can directly compare the extracted allele and the input SNP allele without doing reverse complement, as they are represented on the same strand.