Genetic map

[Ellen R]

Hi Karl, 

There seems to be an issue with my estimated genetic map from a backcross with the estimated map about half the size of the observed genetic map (comparison of observed and estimated map attached). We thought it might be using F2 recombination probabilities but it doesn't look like that's a configuration option in the est.map() function. I tried using qtl2::est_map but don't see a way to plot that in qtl2. Am I missing something with qtl::plotMap() or is there a way to plot genetic maps in Rqtl2?

Thanks, 

Ellen 

[Karl Broman]

I would use summary() on the cross object, to ensure that it's being treated as a backcross ("bc") rather than an intercross ("f2").

R/qtl2 doesn't have a function for plotting maps, but the map objects are the same format as in R/qtl1, so you can use the R/qtl1 function, qtl::plotMap().

karl

[Ellen R]

The estimated map produced by Rqtl2 appears the same. And the cross object is being treated as a backcross. Do you have any other ideas or does this even seem like a problem? 

Ellen 

[Karl Broman]

I can't think of what else might be causing this, but yes for sure, if they were my data, I'd want to figure out the cause.

karl

[Karl Broman]

I think I'd start by picking a chromosome (say chr 1) and looking at the estimated number of crossovers in each individual. The average number of crossovers should correspond to the estimated length in Morgans.

karl

[Ellen R]

Thanks, I wasn't sure where to start so this is helpful. Appreciate the assistance. 

Ellen