Projects: GENIQUEST : Using Rqtl to generate simulated cross data
This page last changed on May 22, 2008 by stepheneb.
Attached zip file: generatemap.zip, has some examples of generating simulated data in R.
You can see in the command:
fake4 <- sim.cross(map1, type="f2", n.ind=400, model = rbind(c(1,45,0.6,0.6),c(2,20,0.5,-0.5)))
In the model you have c(1,45, 0.6, 0.6) the 1 is the chromosome number, 45 is the cM position you want the QTL at and the 0.6 and 0.6 are the effect size. You can make the peak tall/narrow, or sharp/wide by changing these last two numbers.
The trick I used was to make a "cross" file called testmap1.csv. This file has the markers and their location but only fake phenotype data and fake genetic info for three rows. This is an easy file to create for any markers we want to use. I then use the pull.map function to take the map out.
I have include some example sim data in an R script called "fakedatascript.R" in the zip file with the files to run the examples. This will generate a main scan plot and the confidence intervals for the QTL.
|Document generated by Confluence on Jan 27, 2014 16:41|