This is a test version of Biostars. For the public version, visit https://www.biostars.org.
long reads Pacbio SNP calling

Hi all,

I understand that with Pacbio error rate (~15%), it is not really suitable for SNP calling.

This is maybe a naive question, but I was wondering if we have, for example, a really high coverage sequencing of a bacteria (>200X), wouldn't it make it possible to call SNP anyway?

If so, what would be the most cleaver way to do that? Try to do the "classic" way, align to a reference genome and detect variants (is there any tool doing that?). Or maybe perform a genome assembly, and then align the assembly to the reference ?

Has anyone tried that already?

Thanks in advance for your inputs

pacbio snp
• 6,179 views • link
updated by tjduncan tjduncan • written by guillaume.rbt guillaume.rbt

When you have 200x coverage, it might be easiest to simply generate consensus sequence (reads of insert) and then call variants from those. That way you avoid the problems of high error rates.

• 0 views • link
written by Brian Bushnell Brian Bushnell

Ok that was what I thought, thanks for your help.

• 0 views • link
written by guillaume.rbt guillaume.rbt

2 answers

no worries, with high coverage you can do a decent SNP calling using PacBio

• 0 views • link
written by stolarek.ir stolarek.ir

ok great, which tools would you recommend to do that?

• 0 views • link
written by guillaume.rbt guillaume.rbt

either look at their webpage if they have anything dedicated for that, or use GATK

• 0 views • link
written by stolarek.ir stolarek.ir

If you have 200x coverage of a microbe you should just make a de novo assembly with your data using a long read assembler. The two I would recommend are HGAP4 or Canu. Both of these assemblers include a consensus step and will yield an assembly that is of high enough quality to do SNP based variant calling using your preferred bfx pipeline.

HGAP 4 can be used from the command line by downloading PacBio's SMRT Link analysis suite

or

Canu may be the easier tool to quickly set up and use as it is available in bioconda.

• 0 views • link
written by tjduncan tjduncan

Log in to answer this question.

More posts like this