Question about model with ancestral propulation.

[Vanina Castroagudin]

Dear all,

In the vcf tutorial of the 2020 workshop - see figure below-, there is a model where using "an ancestral population" is suggested to get better estimates among the divergent populations. 

Can this model be used even if we do not have " real data" for this " ancestral population"? I mean, if we include it in the model, but do not have data for it, would Migrate run and calculate its parameter somehow ( I don't think so, but I prefer to ask even at the risk of my question sounding ridiculous). Also, for example, if we had data for one or two individuals that could be considered for this ancestral population (for example, from old herbarium material, ancient collections), could that data be used? I mean, are populations with 1 or 2 individuals valid?

Any help is greatly appreciated.

Vanina

Vanina L. Castroagudín, PhD

Visiting Research Scientist 

Mycology & Nematology Genetic Diversity & Biology Laboratory

USDA - ARS

10300 Baltimore Avenue, Bldg 10A, Room 227

Beltsville, MD 20705

[Peter Beerli]

Vanina,

in fact, it is possible to estimate parameters of population of which you have no sample, but do not expect that the estimates are great ;-)

if you have an ancestral population add it to the datafile with a single line (you alsow will need to adjust the number of populations on the the first line)

0 ancestorpopulation

if you have data for such a population then you would need to input also the sampling date and this becomes somewhat cumbersome and will need also more testing with the v4 version. If you have such data already let me know and I will look into this. Even a single individual may help to improve some parameters but for  better population size estimation you will need to have at least 2 individuals for each population.

Peter

On Oct 7, 2021, at 4:29 PM, Vanina Castroagudin <v.castr...@gmail.com> wrote:

Dear all,

In the vcf tutorial of the 2020 workshop - see figure below-, there is a model where using "an ancestral population" is suggested to get better estimates among the divergent populations. 

Can this model be used even if we do not have " real data" for this " ancestral population"? I mean, if we include it in the model, but do not have data for it, would Migrate run and calculate its parameter somehow ( I don't think so, but I prefer to ask even at the risk of my question sounding ridiculous). Also, for example, if we had data for one or two individuals that could be considered for this ancestral population (for example, from old herbarium material, ancient collections), could that data be used? I mean, are populations with 1 or 2 individuals valid?

<Screen Shot 2021-10-07 at 4.25.53 PM.png>

Any help is greatly appreciated.

Vanina

Vanina L. Castroagudín, PhD

Visiting Research Scientist 

Mycology & Nematology Genetic Diversity & Biology Laboratory

USDA - ARS

10300 Baltimore Avenue, Bldg 10A, Room 227

Beltsville, MD 20705

--
You received this message because you are subscribed to the Google Groups "migrate-support" group.
To unsubscribe from this group and stop receiving emails from it, send an email to migrate-suppo...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/migrate-support/ef81b69e-10ed-486b-9aa4-086f2443db09n%40googlegroups.com.

[Vanina Castroagudin]

Hello Dr Beerli,

Thanks for your prompt reply.

For the ancient population, I do have SSR data from one herbarium samples, which is 200 years old and of an Intermedium-old pop from the 70s (4 individual) in addition to my 3 contemporary populations. I also have metadata about collection time and place (aproximated gps coordinates).

Would this work? I have no idea how to built that parmfile….

Thanks,

Vanina

You received this message because you are subscribed to a topic in the Google Groups "migrate-support" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/migrate-support/ZAkaf6-xeLE/unsubscribe.
To unsubscribe from this group and all its topics, send an email to migrate-suppo...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/migrate-support/CF41B106-4898-430E-BF3D-A83EF7CDA356%40gmail.com.

--

Vanina L. Castroagudín, PhD
Visiting Research Scientist
Mycology & Nematology Genetic Diversity & Biology Laboratory

USDA - ARS
10300 Baltimore Avenue, Bldg 10A, Room 227

Beltsville, MD 20705

Cell: (479) 313-3780

Vanina.Ca...@usda.gov
v.castr...@gmail.com 

[Peter Beerli]

The ancestral material most likely will need to be DNA and not msats. If these are annuals then this is about 200 generations which is not that much in terms of different mutations. As a first guess I would just put them as the same as all the current material, I would think that the error will be tiny. For the seventies sample I do not think that the dated sample approach will help at all, just treat them as concurrent with the other sample.

Perhaps next week I can give you an answer about the dated samples with msats

Peter

To view this discussion on the web visit https://groups.google.com/d/msgid/migrate-support/CAOX2sr3nqsEvTDHH0qQCDksa_qEyXK%3D07hHoa3%2B8c9nDENJn%3Dg%40mail.gmail.com.