Impact of Switch Error on Population Divergence Time Estimation

[biozzq]

Dear @stschiff

To evaluate how switch errors affect divergence time estimation, I used the multihetsep_switcherrors.py tool from msmc-tools to process multihetsep files. I set the length parameter to 100 (corresponding to ~1% switch error rate), 50 (~2%), 20 (~5%), and 10 (~10%), respectively, and examined whether the estimated divergence times changed significantly compared to the raw estimation. However, I found that the results were quite similar across these settings and did not differ significantly from the raw estimation.

I am uncertain whether my approach is correct. Could you please provide guidance on whether this is the expected behavior or if there might be an issue with my implementation?

Thank you for your support!

Best regards,
Zhuqing Zheng

[stschiff]

Dear Zhuqing Zheng,

I have not used or maintained that script in more than 10 years. It's pretty straight forward, though, and you should be able to convince yourself that it works by inspecting the multihetsep files and checking that indeed some parts have switched phases.

Regarding the results, I don't really know what to expect. I agree that it's a bit suprising that the rate doesn't seem to matter...