recombination rate?

[Jean]

Hi Joao,

The popABC manual says "The simulation method models each locus as
having segregated independently over time" because the IM model
assumes free recombination between loci.

Yet, the program takes priors on recombination rates.
This seems to contradict the assumption.

Thanks for your help.

Jean

[Joao Sollari Lopes]

Hi Jean,

You are right. How the recombination works is not very clear in the
User-guide. Here how it is:

- For sequence data (sets of SNP's)
By default the sites within a sequence are considered linked, but each
sequence segregates independently.
If you set a recombination rate the recombination events will act
within each sequence (i.e. between sites).

- For microsatellites
By default each locus segregates independently.
If you set a recombination rate, two things can happen: if you set the
recombination rate to be zero, all the loci will be linked; if you set
it to be more than zero then the recombination events will act between
loci.

Hope it is more clear now.

Best,
Joao

[Jean]

Thanks for you reply, but something isn't clear to me yet.

For microsats: you say that "by default each locus segregates
independently". How do I specify that default setting? The prior file
can take the following options (from memory, because I can't use the
software now) : 0 (no recombination), 1 (lognormal prior) and 2
(normal prior).
Am I mistaken and is "0" the default, free-recombination setting?

Jean

On Feb 16, 7:37 pm, Joao Sollari Lopes <j.sollari.lo...@gmail.com>
wrote:

[Joao Sollari Lopes]

Yep, that's correct.

If you leave it as 0, they segregate independently. If you choose 1 or
2 and set the priors to be a point value of 0, they will be linked.
Any other value (> 0) that you set and recombination will act at the
linked microsat.

Joao

[Jean]

Thanks Joao. I'll use option 0.

Jean