DFE for MusMus

[peterdfields]

@igronau I would like to add a DFE for Mus musculus to stdpopsim as we corresponded a bit about before. It looks like there might be reasonable estimates for a couple different kinds of sites from here and here. Given there are a couple of different DFEs one could implement, would it be better to get started with one or the other type or is it easier to submit the different configurations all at once?

[igronau]

I would do them one at a time, and start with whichever looks more straightforward

[peterdfields]

@igronau So I think it might be relatively straightforward at least for a deleterious gamma DFE due in part to the way SLiM was already used in the second study linked above. Specifically, specific values for the Mus musculus subsp castaneous DFE were estimated using polyDFE and summarized in the supplementary materials here. The specific values used in the simulations for CDS, CNE, and UTR in Booker et al. 2021 are described here. In the manuscript they state that they assume a N_e of 42000, so I think simply dividing the S_d value by this value, taking the b value for the gamma shape, would give us a similar set of settings as the Gamma_H17 from DroMel?

[igronau]

I think so. I'm not a huge expert in all the details of modeling the DFE. I suggest that you model this using your best understanding, and then ping me to QC it. I can consult others to make sure this was done appropriately.

[peterdfields]

@igronau okay, I've started the branch. It looks like the recombination map still hasn't been pulled into main. Is there anything I still need to do to allow that pull request to get through? Should I wait on this before starting another pull request?

[igronau]

I think you can work on this regardless of the status of the recombination map. You might want to ping @andrewkern to check what's the status of the recombination map.

[peterdfields]

Do you mean send an email or is there some other way of getting into contact? I think I tried the @ on the issue before but didn't get a response.

[igronau]

Then, maybe send a direct email to [email protected] to figure this out.

[peterdfields]

okay, email sent.

[peterdfields]

@igronau No luck on the map upload so far. Maybe the hold up has to do with the cache problem on AWS that's being worked out?

[andrewkern]

sorry for the holdup-- I just uploaded the map to MusMus on AWS