Clipping primer after alignment

Question

[finswimmer]

Hello everyone,

we are doing targeted sequencing using a Qiagen Kit. During library prep the dna is fragmented and tagged with umi and a universal primer sequence at one end of the strand. So for the enrichment only one specific primer per amplicon is needed.

I would like to clip the primer sequences after alignment. The tools I have found always need the positions or sequence of the primer pair that were used. But in my case I don't have a traditional primer pair for a amplicon.

I have the position, the orientation and the sequence of the primers in the kit.

Is anyone aware of tool that can clip my reads after alignment?

Thanks a lot.

fin swimmer

[gb]

Under the header "Example UMI extraction" there is maybe something usefull https://umi-tools.readthedocs.io/en/latest/QUICK_START.html#step-5-deduplication

[finswimmer]

Hello gb ,

thanks for your reply. Could you please explain why you think that there is something useful for me? Unfortunately I couldn't find anything :(

fin swimmer

[gb]

oh, I thought maybe because of this text:

The case below could be handled by umi_tools dedup by specifiying –bc-pattern=NNXXXXNN but we include it here as an example of how the fastq should be formatted following extraction: UMI is bases 3-7, bases 1-2 and 7-8 are the sample barcode and need to be removed

[michael.ante]

Hi finswimmer,

A possible solution is to convert your BAM file into BED format and use bedtools subtract with your primer BED to get clipped locations. Using bedtools getfasta would get you the clipped sequences.

Could you please tell, why you keep the primer sequences in the reads during alignment?

Cheers,

Michael

[finswimmer]

Hi Michael,

thanks for your suggestion. Get the positions to clip is the easy part, but doing the clipping and clip also overlapping bases from the mate strand is more complicating. I've started to write my own tool for it. Let's the if I can finish it before someone presents an already existing tool.

Could you please tell, why you keep the primer sequences in the reads during alignment?

This improves the mapping. Longer reads are better to map. Also the primer sequences are (or should be) specific, this improves the mapping as well.

fin swimmer

[dlaehnemann]

Hi fin swimmer,

any luck with the QIAseq primer trimming? We want to do exactly this and also haven't found a tool to do this. So we would go and implement something for it as part of rust-bio-tools, but any previous work would be interesting.

Best regards and thanks, David