QTL mapping using physical distance

[fishlike]

Can I use physical distance, rather than genetic distance, for QTL
mapping? I'm asking this because I couldn't estimate the genetic
distance for all the markers due to non-inbred crossing scheme.

Thank you

[Karl Broman]

You could just do analysis of variance at the markers, which doesn't rely on the map, but if you wish to do interval mapping, you'll need a genetic map.

karl

> --
> You received this message because you are subscribed to the Google Groups "R/qtl discussion" group.
> To post to this group, send email to rqtl...@googlegroups.com.
> To unsubscribe from this group, send email to rqtl-disc+...@googlegroups.com.
> For more options, visit this group at http://groups.google.com/group/rqtl-disc?hl=en.
>

[Haili Zhang]

Say, if I have a set of markers whose genetic distance and physical locations are both known, and I have another set markers with only physical location known, is there ways to combine these 2 sets of markers and estimate the genetic distance of all the markers? (such as, based on local recombination rate, estimate how many cM is equivalent to 1Mb?)

 

A separate question, can I somehow use physical location to guide constructing the genetic map;  Or making genetic map can not make any use of known physical genome information, other than ajusingt parameters to make the genetic map consistent with known genome? 

 

Thank you.

[Karl Broman]

You definitely use the physical map to guide the construction of the genetic map. If you know the physical locations of all markers, you simply use est.map() to estimate the inter-marker distances to get the genetic map.

I would either ignore the previous genetic distances (available on just a portion of the markers) and start with Mbp positions of markers, or I would use linear interpolation to get genetic locations for markers for which you have Mbp but not cM positions.

karl

[Haili Zhang]

How do I add physical locations to the .csv file? I'm reading your maual of genetic map construction and R/qtl, still don't have a clue how to do this?

 

My understanding is that I need to start with a file similar as "mapthis";  get the genetic location in cM for every marker used for the map; interpolate the genetic locations for those markers I only have Mbp; and add the genetic locations for all the markers in a file similar as "sug" or "hyper"; and start QTL mapping from there.

 

Thanks

[Karl Broman]

the csv file would include just one set of marker locations; include them in the third row.  look at the sample files at http://www.rqtl.org/sampledata/ and also Ch 2 of the R/qtl book (http://www.rqtl.org/book)

karl