search

stschiff / msmc2

GNU General Public License v3.0
53 stars 9 forks source link

about RCCR value > 1 #62

Open cnxiaobo94 opened 4 months ago

cnxiaobo94 commented 4 months ago

Dear @stschiff,

I have a dataset that consists of three populations as (pop1, (pop2, pop3)). After running msmc2, I have three paired RCCR values, but I met some problems when plotting them. As the picture shows below, some values of time points are out of rates 0-1. I can observe the divergence times of pop1&2 (orange line) and pop1&3 (green line), but the divergence time (blue line) between pop2 and pop3 is hard to know. I'd appreciate it if you could help with this problem. rccr_1

Best, Bo Xiao

stschiff commented 3 months ago

Looks to me like pop2 and pop3 are hardly diverged at all. Have you checked with other statistics, like FST, or looked at whether the population sizes start to differ somewhere?

cnxiaobo94 commented 2 months ago

Looks to me like pop2 and pop3 are hardly diverged at all. Have you checked with other statistics, like FST, or looked at whether the population sizes start to differ somewhere? Thanks for your reply! Sorry for this denied message. I checked the popo]ulations siez of both MSMC and PSMC (as below). The divergence time of pop2 and pop3 is calculated as between 100-200 ka. The weighted Fst estimate of pop2&3 is 0.25, while that of pop1&2 (pop1&3) is 0.56. image image