Inquiry: Estimating Wild–Cultivated Divergence by RCCR

[doubleHwithT]

Dear Dr. Schiffels,
I hope this message finds you well.
Thank you and your team for developing such efficient tool for reconstructing population histories!

I would like to ask for advice on using MSMC2 to calculate the RCCR for inferring the divergence time between two subpopulations.

First, I have run MSMC2 on each population (4 samples, 8 haplotypes each group), Here are the MSMC2 commands for the two subpopulations :
pop1:
msmc2 -t 30
-I 0,1,2,3,4,5,6,7
-p 12+251+1*2
-o pop1_singal.out
pop1_and_pop2_phased_Chr01.multihetsep.txt … pop1_and_pop2_phased_Chr13.multihetsep.txt

pop2:
msmc2 -t 30
-I 8,9,10,11,12,13,14,15
-p 12+251+1*2
-o pop2_singal.out
pop1_and_pop2_phased_Chr01.multihetsep.txt … pop1_and_pop2_phased_Chr13.multihetsep.txt

My study focuses on one wild subpopulation and its corresponding cultivated subpopulation. For each group, I selected 2 samples (4 haplotypes) to perform an RCCR analysis. The command I used is:

msmc2 -t 34
-I 0-8,0-9,0-10,0-11,1-8,1-9,1-10,1-11,2-8,2-9,2-10,2-11,3-8,3-9,3-10,3-11
-s -p 12+151+1*2
-o pop1_pop2_compar.out
pop1_and_pop2_phased_Chr01.multihetsep.txt ... pop1_and_pop2_phased_Chr13.multihetsep.txt

msmc-tools/combineCrossCoal.py pop1_pop2_compar.out pop1_singal.out pop2_singal.out > pop1_pop2_combine_cross.final.txt

The RCCR results are shown in the figure:

The RCCR curve indicates that RCCR = 0.5 corresponds to approximately 60,000 years ago. However, for a wild vs. domesticated plant comparison, this seems too early. I have seen other studies reporting an RCCR-derived divergence time of around 10,000 years ago for wild/cultivated plants, which better matches the known timeline of domestication by humans.

Could you please help me understand how to correctly interpret the RCCR results? If I want to estimate the true divergence time between wild and cultivated groups, do I need to adjust any analysis parameters?

Thank you very much for your help, and I look forward to your reply.

Bruce

[stschiff]

Hi. I can only say that this RCCR looks quite good (for example, it is nearly monotonic). I don't know anything about Rice evolution, so cannot comment on the interpretation. Have you checked the population size curves? Do they look reasonable? Perhaps the domesticated one has such an extreme population size bottleneck that this affects the RCCR in some weird way? Also, have you tried other methods to measure divergence, for example Site Frequency methods like MOMI2?

[doubleHwithT]

Dear Dr. Schiffels,

Thank you for your reply.

I’ll go back and review my MSMC2 workflow to see if anything went wrong. Over the past few days I’ve been running MOMI2, but the divergence time estimates have been very sensitive to the thresholds I set. I’ll keep experimenting with MOMI2 over the next few days to see what results I get.

Best regards

Bruce

[doubleHwithT]

Dear Dr. Schiffels,

I hope you’re well.

I have checked the effective population size estimates for my wild and cultivated groups, and the results are shown in the attached figure.

I am currently investigating inter‐population introgression and have discovered that the cultivated group harbors substantial gene flow from other species. For example, when I compared species A wild vs. A cultivated, I had expected a divergence time of around 10,000 years ago, but the RCCR‐inferred time was approximately 40,000 years ago—much earlier than anticipated. My recent analyses suggest that A cultivar has received a large amount of introgression from species B, even though A species and B species diverged long ago. I suspect that this external introgression is biasing the RCCR estimates.

I am considering using MOMI2 to fit a more appropriate demographic model and recalculate the divergence time.

Thank you very much for your guidance.

Best regards,

Bruce

[stschiff]

Sounds good.