Population selections

[cnxiaobo94]

trafficstars

Hi, I'm doing some population demographical analysis among three populations, and I have two questions I'd like to ask for help.

1. There is only one individual in population A, and whether it can be analyzed?
2. If the previous question is yes, can the three populations (A, B, C) be simulated at the same time (in one command line), or do simulations in pairs (A-B, A-C, B-C)? Looking forward to your reply! Thanks!

[tcztzy]

1. As far as I know, you can't analyze "population" with only one individual, it doesn't make sense, "I suggest you don't actually run this because we won't need this single-sample processing". Actually, MSMC2 recommends you provide trio (three individuals, parents and their offspring).
2. msmc2 could take more than one .multihetsep.txt file. You can write your own shell script to run cross-population estimate.
3. msmc-tools provided a friendly guide for very beginner, if you still have HowTo problems, maybe that repo is a good place to open issue.

[stschiff]

Hi, I'm doing some population demographical analysis among three populations, and I have two questions I'd like to ask for help.

1. There is only one individual in population A, and whether it can be analyzed?

This would then be like in PSMC. That is possible, and would give you effective population size estimates for the population from which this sample is from.

2. If the previous question is yes, can the three populations (A, B, C) be simulated at the same time (in one command line), or do simulations in pairs (A-B, A-C, B-C)?

I don't know what you mean by "simulate". If you mean "analyse", then no, MSMC/MSMC2 cannot analyse three populations at the same time. But you could estimate separately:

• coalescence rates within populations A, B and C
• cross-coalescence rates between populations A-B, A-C and B-C.

[tcztzy]

Hi, I'm doing some population demographical analysis among three populations, and I have two questions I'd like to ask for help.

1. There is only one individual in population A, and whether it can be analyzed?

This would then be like in PSMC. That is possible, and would give you effective population size estimates for the population from which this sample is from.

2. If the previous question is yes, can the three populations (A, B, C) be simulated at the same time (in one command line), or do simulations in pairs (A-B, A-C, B-C)?

I don't know what you mean by "simulate". If you mean "analyse", then no, MSMC/MSMC2 cannot analyse three populations at the same time. But you could estimate separately:

• coalescence rates within populations A, B and C
• cross-coalescence rates between populations A-B, A-C and B-C.

If the individual is homozygote (like rice and other self-pollinated plant), can it still be analyzed? I tried but it seems Powell minimization exceeding maximum iterations.

[stschiff]

No, you can't analyze a homozygous genome alone. You can of course treat a homozygous genome as a single haplotype and combine it with other such genomes to create something like a pseudo-diploid to estimate coalescence rates between them.

[tcztzy]

No, you can't analyze a homozygous genome alone. You can of course treat a homozygous genome as a single haplotype and combine it with other such genomes to create something like a pseudo-diploid to estimate coalescence rates between them.

Thank you for your patient, it helps me.