search

nf-core / eager

A fully reproducible and state-of-the-art ancient DNA analysis pipeline
https://nf-co.re/eager
MIT License
141 stars 80 forks source link

Add BowTie2 support #257

Closed jfy133 closed 4 years ago

jfy133 commented 4 years ago

Is your feature request related to a problem? Please describe. This was previously in EAGER1, but also has benefit of "better for investigating absent gene regions since it can split the reads when mapping." (requested by Ash V. in CPH)

jfy133 commented 4 years ago

Recommended maping settings for aDNA: https://academic.oup.com/mbe/article/35/5/1120/4844088

Additional for UDG treatment and looking for methylation: https://www.frontiersin.org/articles/10.3389/fevo.2020.00105/full#B4

jfy133 commented 4 years ago

Ok, changed my mind on this one. We should include it in 2.2 because it would be important for eager1 parity and also given the Poullet and Orlando paper, would make it more relevent.

I will try to contact the authors and get their recommendations.

apeltzer commented 4 years ago

Docker building here in the repository could also be added then, if we anyways play around with the dev container adding bowtie2 now too

apeltzer commented 4 years ago

Bowtie2 dependency is added too. Remaining would be to allow for alternative mapper instead of bwa ;-)

apeltzer commented 4 years ago

This study reveals that the Bowtie2 mapping conditions recommended (sensitive option, end-to-end mode) returns with larger numbers of read alignments that increase the number of genomic windows available for inference as well as their sequence coverage, which improves accuracy of the predictions. This has important consequences for the nascent field of ancient epigenomics, which, to the best of our knowledge, based all previous predictions on BWA DNA alignments.

https://www.frontiersin.org/articles/10.3389/fevo.2020.00105/full

So we should enable at least that mode if possible :-)