Models

When we speak of models for analysis, we mean to specify the markers used as cofactors in composite interval mapping. There are presently six models for analysis.

1. Use all the markers to control for the genetic background. This is model 1 from zeng@94.

2. Use all unlinked markers to control for the genetic background. This is model 2 from zeng@94.

3. Don't use any markers to control for the genetic background. This is also known as interval mapping and is the same as Lander and Botstein's method [Lander and BotsteinLander and Botstein1989].

4. This is an ad-hoc model. One marker from each chromosome (except for the chromosome on which we are testing) is used to control for the genetic background. The results of LRmapqtl are scanned, and the marker that showed the highest test statistic from each chromosome is used.

5. This is another ad-hoc model. Two markers from each chromosome are used to control for the genetic background. They are the top two markers as determined by LRmapqtl. In addition, all the other markers on the chromosome of the test position that are more than 10 cM away from the flanking markers are also thrown in. It may be ad-hoc, but tends to work best at this time. The value of 10 centimorgans can be changed with the -w option.

6. Model six will be explained in the next subsection.

7. Model seven requires the results of a prior run of Zmapqtl and Eqtl. Initially, the user may want to run Zmapqtl with interval mapping, summarize the positions and effects of that analysis using Eqtl, and then use those estimates as the covariates in the regression model. Virtual markers are created at the best estimates for the positions of the QTLs.

8. Model eight is similar to model seven except that instead of using virtual markers, the nearest flanking markers to the putative QTL are used as cofactors. As in model seven, Eqtl is run after Zmapqtl, and a table of markers with arbitrary ranks are written to the SRmapqtl.out file. You should set the number of background parameters to more than the number of QTL identified by Eqtl.