Fix covariance parameters

Check this box to estimate the covariance parameters, for the K matrix and/or additional random effects that can be specified, in a model including all non-genetic effects, and use these estimates in the association testing.

Having these parameters fixed eliminates the need for the covariance parameters to be re-computed for every SNP mixed model fit and saves computation time. This option is based on the population parameters previously determined approach (P3D) that was proposed concurrently with the optimized K matrix compression method in Zhang et al. (2010)¹.

You should not check this option if you are already using the PARMS statement generated by K Matrix Compression.

Note: This option is ignored when no K Matrix Square Root Variables² and no Additional Fixed Effects are specified.

Note: You should not select this option when either SATTERTHWAITE or KENWARDROGER is selected as the Denominator Degrees of Freedom Method (on the Model Variables tab).

---

1

Zhang, Z., E. Erzoz, et al. (2010). Mixed linear model approach adapted for genome-wide association studies. Nature Genetics 42: 355-360.

2

This parameter is not found in any Workflow process dialog, but is found in some individual underlying Genetics process dialogs.