lodint using cim results

[vega....@gmail.com]

Karl,

I was reading the documentation for lodint in R/qtl. It suggests that the user inputs in the results argument the output of scanone or qtl object from refineqtl.

I noticed that the function will work with the output of cim. Is there anything wrong with using the output of cim as the results argument?

# Estimate LOD support inverval with a drop of 1.5
lw_int_pop1 <- lodint(cim_qtl_results1$length_width_ratio, chr = 4,
                 expandtomarkers = TRUE,
                 drop =  1.5)

> # Print 1.5 LOD support interval
> lw_int_pop1
              chr      pos       lod
CHR4_8401383    4 37.71007 3.8160374
CHR4_10113829   4 39.87007 5.5211651
CHR4_63432223   4 47.77016 0.3225328


When I replace the results argument in lodint from cim_qtl_results1$length_width_ratio with a QTL object obtained using refineqtl, the QTL position changes significantly.

# Refine QTL object
lw_qtl1_pop1 <- refineqtl(M1, qtl = lw_qtl1_pop1, method = "hk", keeplodprofile = TRUE)

# Estimate LOD support inverval with a drop of 1.5
lw_int_pop1 <- lodint(lw_qtl_pop2 chr = 4,
                 expandtomarkers = TRUE,
                 drop =  1.5)

> # Print 1.5 LOD support interval
> lw_int_pop1
              chr      pos        lod
CHR4_149657     4  0.00000 0.04933776
CHR4_72150974   4 61.89021 0.43241092
CHR4_73098082   4 64.42021 0.28509443

Please advice thank you.

Andrey Vega Alfaro
Dep. of Plant and Agroecosystem Sciences
Univ. of Wisconsin-Madison.

[Karl Broman]

Yes, the output of cim() has the same format as the output of scanone(), and so you can use lodint() with it in the same way.

But the results are going to be quite different from the use of refineqtl() and related functions, particularly because of the use of windows around selected markers in cim().

karl

[Andrey Vega]

Thank you! 
is there a better more precise approach? I plan to use lodint with the output of cim as I think those are more sensible results and correspond to previous qtl hits. 

--
You received this message because you are subscribed to the Google Groups "R/qtl discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to rqtl-disc+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/rqtl-disc/afb934b1-af01-4793-81d0-9ae3431a1ccdn%40googlegroups.com.