Nm and mutation rate

[Rie]

Hello.

I would like to calculate Nm (number of migrants per generation) using data set of mtDNA COI sequences.

But I calculated  without  "mutation rate" and "number of generation per year" (Datatype and Data Specific Options).

In this case, can I calculate Nm using θ and M?

And I have a question.

I would like to use the mutation rate 3.54%/ 1000000 year, 

what should I input the values on the Datatype and Data Specific Options of "mutation rate" ?

I'd appreciate it if you could answer about these.

  

[Peter Beerli]

Nm = theta *M/4.     (For diploid)

The mutation rate entries int the menu are for dated serial samples

P

Peter

(sent from by phone)

--
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 post to this group, send email to migrate...@googlegroups.com.
Visit this group at http://groups.google.com/group/migrate-support.
For more options, visit https://groups.google.com/d/optout.

[Rie]

Thank you for your message.

And I would like to  know how to calculate the mutation rate (μ)??

Because I have to explain above for my supervisor and reviewers.

They asked a question "Did you specify a mutation rate? What was it and how was it obtained?"

I have to answer about this.

[Rie]

Peter 

Thank you for your comments!!

And I would like to know how　to calculate μ ?

Please tell me, this calculation method or　index to calculate μ.

Because I have to explain above for my supervisor　and 　reviewer.

I'd appreciate it if you could answer about these.

Rie

2015年7月10日金曜日 20時33分07秒 UTC+9 Rie:

[Peter Beerli]

Rie,

> They asked a question “Did you specify a mutation rate?

Answer: “No”
> What was it and how was it obtained?”
Answer: “?” [given that you did not use a mutation rate — the followup question is unanswerable]

Explanation:
Migrate-n estimates (the manual discusses this)
- mutation-scaled population size Theta: the program returns a value for example 3.2 for msat, or 0.0098 for DNA data. This allows comparison with other populations that were also run, if you have two populations, for example, Theta_pop1=0.0098, Theta_pop2=0.0134 then you can say that pop2 is 0.0143/0.0098=1.45x bigger than pop1. If you want to know absolute numbers, e.g. effective population size Ne, then you need to know that Theta=4*Ne*mu, this of course is only possible if you know mu from somewhere (for example the literature; you may want to read
Whales before whaling in the North Atlantic J. Roman and SR Palumbi Science (New York, NY), {JUL 25} 2003 vol. {301} ({5632}) pp. {508-510}
for an example of the conversion of Theta to Ne.


- mutation-scaled immigration rate M, which again is a value that may look like this for the immigration rate from pop2 to pop1 M_pop2->pop1=132.0 and again you can compare with other populations, it is defined as m/mu where m is the immigration rate and mu is the mutation rate per generation (remember: for msats it is per locus, for DNA it is per site). To calculate the number of immigrants Nm you can do the following: Theta*M = 4*Ne*mu * m/mu = 4Nm, no need to know the mutation rate because it cancels. you will need to make sure that you use the correct values, because you need the population size of the population the migrants migrate into: you need to calculate for example 4Nm = Theta_pop1 * M_pop2->pop1 = 0.0098 * 132 = 1.29 ==> Nm=1.29/4

If you have mtDNA instead of the 4 use 2.

all clear?
Peter

[Rogério MG]

Hi Peter,

 

Do we obtain these values (0.0098, 0.0134 and 132.0) from the “posterior distribution table” (Figure 25 from the manual)?

Rogerio

[njgi...@gmail.com]

Hello, Rie

Have you understood how to caculate Nm? I have some questions about the formula.
(1) In the results of Posterior distribution table, we also got several percentages in the first line. Which percentage I shoud select for caculating Nm?
(2) Generally, for nuclear data it x = 4, for haploid data its x = 2, and for mtDNA in vertebrates with female-only transition it is x = 1 (Accrding manual ). But yours sequences are COI, why Peter tell you X is 4 instead of 2 ?

Thank you for your answer firstly!
Gina Xu

在 2015年7月10日星期五 UTC+8下午7:33:07，Rie写道：

[Peter Beerli]

Gina, and Rie

I answered to Rie with x=4 because the COI remark was two emails before — I rarely read old message when I answer because there is only so much time

in a day to answer questions. If the data is mtDNA then it depends on the organisms, but most likely it will be x=1 because your species has a sex ratio of 1:1 and then the mtDNA is haploid. 

To recapitulate:

x * Ne_x* m_y->x = Theta_x*M_y->x

for diploid x=4

for haploid x=2

for haploid mtDNA + sex ratio 1:1 AND YOUR SPECIES DOES NOT CHANGE SEX DURING THEIR LIFETIME x=1

[I use caps because many fish do exactly that thus technically every individual can contribute mtDNA to the next generation]

And one more:

migrate does NOT estimate mutation rate with contemporary data, therefore you cannot enter a mutation rate value and get answers, only with date samples that 

may be possible,

Peter

P.S. the google group has _many_ answer on the questions about Nm, how to multiply, what inheritance scalar to use, and the problems with mutation rates