This is a test version of Biostars. For the public version, visit https://www.biostars.org.
How to do a pairwise alignment and convert the result to a VCF file?

Hi,

I need to do a lot of pairwise alignments and obtain a VCF file of each alignment. The purpose of this is to find the mutations between different ORFs of a reference genome and this ORFs in other genomes (one by one). Is there some combination of software that can do this? I am not sure that this can be done with MAFFT or muscle, since they are multiple aligner, and I am not sure that this can be done with a mapper like minimap2 or BWA, since mappers works with reads.

Maybe I could use MAFFT and build a parser in Python to see the mutations in each alignment, but maybe there is a program (or combination of program) that does this more easily.

Thanks,

vcf genomics mutations
• 3,554 views • link
updated by johnston.mike.j • johnston.mike.j • written by arturo.marin • arturo.marin

1 answer

You could use minimap2 and paftools, some inspiration can be found here: https://github.com/lh3/CHM-eval/blob/master/dip-call/README.md

• 0 views • link
written by WouterDeCoster WouterDeCoster

Following the Minimap2 cookbook:

minimap2 -cx asm10 --cs "$ref_fasta" "$asm_fasta" \
  | sort -k6,6 -k8,8n \
  | paftools.js call -L 1000 -f "$ref_fasta" - > "$out_vcf"

Breaking this down:

Align using minimap2. Include the options -c: output CIGAR -x asm10: Preset parameter for alignment between assemblies assuming <1% sequence divergence. (Could use asm5 or asm20 if your sequences are more/less similar.) --cs: output cs tag

Sort the PAF output

Redirect to paftools.js call. By default, alignments less than L=50000 are not included for variant calling. Set L to lower value if your assembly is smaller than this. Provide reference fasta sequence with -f.

• 0 views • link
written by johnston.mike.j • johnston.mike.j

Log in to answer this question.

More posts like this