search

hardingnj / xpclr

Code to compute the XP-CLR statistic to infer natural selection
MIT License
85 stars 26 forks source link

Input file and genetic map requirement #61

Open hungweichen0327 opened 3 years ago

hungweichen0327 commented 3 years ago

Dear all,

I am a beginner of using this powerful software. After reading the manual and other related issue, I would like to confirm about the input file format and input requirement.

  1. For the input file, I can just use vcf file? And I also need to set the “-rrate” for the fixed recombination rate along the genome? This means I don't need genetic map data for the analysis. Is that right?

  2. So far I don't have genetic map data. I know genetic map data is the key part for XP-CLR analysis. I am not sure whether using vcf file without genetic map data can't have precise identification of selective sweeps. Do you have any suggestions if I only have vcf file?

hardingnj commented 3 years ago

Hi @hungweichen0327

  1. Yes- that is correct. If you use a VCF you can skip the map requirement using -rrate
  2. The calculations in xpclr are reasonably robust to recombination rates. It may make a difference on an individual window that is split over a recombination hotspot or coldspot, but other factors such as the availability of SNPs in both populations and population demographic history are likely to be much more important.
hungweichen0327 commented 3 years ago

Hi @hardingnj,

Thank you for the quick and clear reply. For the second answer, you mentioned that "such as the availability of SNPs in both populations and population demographic history are likely to be much more important". Do you mean these two factors may impact the XP-CLR analysis more than having genetic map data? Most paper combined different way to detect the selective sweeps, e.g. XP-CLR, Fst, diversity ratio, CLR. I just think whether I can get reliable result when using XP-CLR with fixed -rrate.