search

nf-core / eager

A fully reproducible and state-of-the-art ancient DNA analysis pipeline
https://nf-co.re/eager
MIT License
134 stars 79 forks source link

GenotypingGalore: add aDNA specific genotypers #451

Open jfy133 opened 4 years ago

jfy133 commented 4 years ago

A bunch of aDNA-aware genotypers have been released relatively recently.

We should add these as possible options

Here is a running list:

More might come, so will keep this for all.

jfy133 commented 4 years ago

Note @sc13-bioinf already has made some work on snpAD:

I made a nextflow pipeline for snpAD so you can copy some of that if you want. The main reason my pipeline sucks is that it builds the substitution model for all chromosomes. You can just do that for one chromosome (or maybe X.Y.MT also if you really want) and then use that. Its in my HOMErepos/nextflow-snpAD but its not a git repo.

But that it doesn't appear avaliable on github, so might be difficult for conda packaging? https://bioinf.eva.mpg.de/snpAD/

apeltzer commented 4 years ago

Its a versioned URL (2.3.4), so one could just make a conda recipe for it - shouldn't be a problem.

jfy133 commented 4 years ago

ANGSD is maybe a better for human shotgun too, according to @TCLamnidis , so should read recenr papers doing it and use those commands

jfy133 commented 3 years ago

Reviewer also requested samtools mpileup + bcftools, so we should include this in this release.

Example commands (from a snakemake pipeline):


" samtools mpileup -q30 -x -s -O -d3000 -f {input.ref} {input.bamA} > {output.pileup} ;" 
    " samtools mpileup -q30 -t SP -d3000 -vf {input.ref} {input.bamA} > {params.vcf_raw} ;"
    " bcftools call -c -Oz -o {output.vcf_strain} {params.vcf_raw} ;"
    " bcftools view -Oz -v snps -q .75 {output.vcf_strain} > {output.variants} ;"
jfy133 commented 3 years ago

I've not really seen most of the tools used tbh (other than mpileup and very occasionally snpAD).

I think mpileup is very common in general so we should at that - but all the others arne't easily added to bioconda, not used, or produce EXTREMELY large output files apparently (snpAD), so I think until those become more widely used we can leave them out. THoughts @apeltzer ?

So 2.4 is add mpileup and then close this issue for now.

apeltzer commented 3 years ago

Agreed