search

wangke16 / MSMC-IM

A tool for estimating time-dependent migration rates based on cross-/within-population coalescent rates from MSMC
16 stars 2 forks source link

How to interpret the result when the inferred rCCR is different from that directly from MSMC2? #7

Open biozzq opened 3 years ago

biozzq commented 3 years ago

Hi @wangke16

After fixing the plot issue, I generated the PDF files for my desinged population pairs; however the results of some pairs are difficult to interpret according to your published paper. For example (like following), the inferred rCCR profile is significant different from the rCCR directly from MSMC2, do you have any comments on this different, and can i use the cumulative migration probability to deduce the separation history under such circumstances? Thank you very much.

image

Best, Zheng zhuqing

wangke16 commented 3 years ago

Hi there,

The inferred rCCR you see here is a native conversion from MSMC-IM internally inferred tMRCA distribution. The equation for calculating becomes numerically unstable for very ancient times (that’s where you observe significant difference in your plot)

In general, I suggest to use the cumulative migration probability (which numerically starts from 0 and reaches 1) for population separation inference. The plateaus you observed in cumulative probability are likely reflections of gene flow/population separation/introgression.

best.

On 30 Jun 2021, at 09:45, biozzq @.***> wrote:

Hi @wangke16 https://github.com/wangke16 After fixing the plot issue, I generated the PDF files for my desinged population pairs; however the results of some pairs are difficult to interpret according to your published paper. For example (like following), the inferred rCCR profile is significant different from the rCCR directly from MSMC2, do you have any comments on this different, and can i use the cumulative migration probability to deduce the separation history under such circumstances? Thank you very much.

https://user-images.githubusercontent.com/15704257/123921954-22726980-d9ba-11eb-97fd-7fe7a119b8fa.png Best, Zheng zhuqing

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/wangke16/MSMC-IM/issues/7, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGLYZY7WUK6WS4ZHTXJMGXLTVLDSDANCNFSM47RYHL3Q.