These are chat archives for dereneaton/ipyrad

27th
Jun 2018
Patricia Lang
@plantricia_twitter
Jun 27 2018 15:33
Hi @isaacovercast and @dereneaton ,
I am using ipyrad for ddRAD data from plant species that don't have a reference genome, and would like to use the consens-loci to build a pseudo-reference that I can map to. Unfortunately, I have not quite been able to figure out which of the output files would be the most convenient to use to get at the 'reference'-versions of my clusters and would love to hear your opinions / suggestions for this. Thanks a lot!
Isaac Overcast
@isaacovercast
Jun 27 2018 16:28
@plantricia_twitter I'm not sure how mapping to a pseudo-reference would be any different than just using the denovo assembled loci from ipyrad, but maybe i'm missing something. If you want a pseudo-reference idea you could just grab 1 sample from each of the loci in the _across/*_catclust.gz file, but then you've already clustered all the samples together at this point anyway. What are you trying to accomplish exactly with the pseudo reference?
Patricia Lang
@plantricia_twitter
Jun 27 2018 16:58
@isaacovercast Thanks for the fast response! Sorry, I should have been clearer. In addition to getting conventional ddRAD data, I used my ddRAD libraries to make baits and capture very short DNA fragments (ancient DNA, cannot be used for making normal ddRAD libraries), ie enrich these short fragments for my RAD loci. This is to have more overlap between the conventional ddRAD and aDNA sequence reads, and be able to compare the two. For the analysis and for example very basic stats, like calculating capture efficiency, I need to map the aDNA reads to a pseudo-reference to get a ratio of mapped/unmapped reads.
As I understand it, the easiest would be to take the sample.loci file, use a custom-script to make a consensus for each locus and use this for a reference-fasta. Like this I would be able to keep the locus-ID as well as the base-positions (to still know the positions of the SNPs found in the sample.vcf). Or is there an easier way that I am missing?
Isaac Overcast
@isaacovercast
Jun 27 2018 18:04
Ooohhhhh, that makes sense. Yeah the way you say with the sample.loci file makes the most sense to me. Sounds cool, good luck!
Patricia Lang
@plantricia_twitter
Jun 27 2018 20:41
Too bad there is no other way – but thanks nevertheless for the confirmation :)