search

hardingnj / xpclr

Code to compute the XP-CLR statistic to infer natural selection
MIT License
90 stars 27 forks source link

Questions about inbreeding species and parameter #73

Open Axolotl233 opened 3 years ago

Axolotl233 commented 3 years ago

Hi community,

I wonder if xpclr method can apply on inbreeding plant species, although i know it was originally designed to dectect the selection of human genome. Besides, i want to know the influence of parameter --ld LDCUTOFF, -L LDCUTOFF and --rrate RRATE, -R RRATE, what will be happen if i change it

all bests, nemo

Axolotl233 commented 3 years ago

Hi, is there any advice?

hardingnj commented 3 years ago

Apologies. I think I would interpret results of highly inbred populations with extreme caution. XP-CLR identifies regions of the genome that have drifted more than you would expect under neutral conditions. The genome of highly inbred species will not behave as a neutral background, there will be much more fixation in a highly inbred population adding a lot of noise.

I would suggest using msprime to simulate some inbred populations to investigate more directly.

--ld cutoff is the threshold at which we exclude SNPs for being in high LD as SNPs need to be independent for the model. --rrate is the recomination rate per base, which affects the windows selected for XPCLR.

Axolotl233 commented 3 years ago

@hardingnj thank you for your advice, i will do some simulate to investigate it.