Adding a genetic map for Apis mellifera

[janaobsteter]

• Name: Apis mellifera
• Common name: honeybee

Other information: Genetic map
Hi! I am trying to add a genetic map to the honeybee, but I've ran in some issues.
So, I have the genetic map, which was done on microsatellites and then lifted onto the current honeybee assembly (this was done by some custom scripts by a collaborator).

However, now I have the positions and the genetic positions, but I don't have the recombination rate (the third column) - and my brain doesn't work so I don't know how to transform it (if I even can?). I do have the constant recombination rates from another paper (by chromosome), but I don't know whether that's ok?

Also, I don't know how to cite the custom scripts for transforming the microsatellite genetic map to SNP based genetic map?

I would appreciate any help on this!
Thanks!

[petrelharp]

Ah, great! Positions and genetic positions are okay: if the positions are pos (in bp) and the genetic positions are map (in cM) then the recombination rate (in cM/Mb) is

rate = 1e6 * np.diff(map) / np.diff(pos)

This will be one number too short, and so you add an extra 0.0 at the end. Is this enough for you to get it in there?

Also, I don't know how to cite the custom scripts for transforming the microsatellite genetic map to SNP based genetic map?

Let's see - so, you're going to cite the original publication, right? How long are the scripts? We do need to figure out how someone might QC this. If they are short you could just stick the scripts in the python file, commented out.

[janaobsteter]

Gotcha, I think I can handle this! Another question - the chromosome names should be the same as in the "genome.py" file (the main name)?

Regarding the script - it's publicly available here on bitbucket, however, it is not mine (https://bitbucket.org/scriptBee/hapmap-pilot/src/main/GeneticMaps/, file create-geneticMap.R).

[janaobsteter]

OK, I think I did it right - I've computed the rate from positions separately for each chromosome with the last value being zero.

for chromosome in range(1, 17):
    chrDF = gmap[gmap.Chromosome == chromosome]
    chrDF = chrDF.sort_values(by = "Position")
    chrDF.loc[:, "Rate"] = list(1e6 * np.diff(chrDF.Map) / np.diff(chrDF.Position)) + [0]
    if chromosome == 1:
        gmapDF = chrDF
    else:
        gmapDF = pd.concat([gmapDF, chrDF])

I've also used the chromosome names used in genome.py.

I've pushed it to my fork here: https://github.com/janaobsteter/stdpopsim/tree/main/stdpopsim/catalog/ApiMel

[petrelharp]

That seems right. Perhaps you could make a pull request?

[janaobsteter]

Created - #1399

[janaobsteter]

OK! I've created the python file - I'm just waiting for some text. But I have another question - should the genetic map provide genetic positions for every base in the reference genome? Thanks!