I'm attempting to run SKAT for the gene based tests. For the null model section, the code is: obj <-SKAT_Null_Model(phenotype ~ covariates_matrix, out_type="D") where phenotype is a numeric binary of 0/1 for case/control status. Does it matter if 0 is case or control and vice versa? Secondly, how exactly do you get the covariates matrix? For gene based tests, what are the covariates? If all I have is patient ID, case/control status and genotype data (0/1/2 or NA), how should I set this up? Thanks, Cian

On Wed, Aug 14, 2013 at 5:05 PM, Cian Murphy <cian.murphy.11 at="" ucl.ac.uk=""> wrote: > I'm attempting to run SKAT for the gene based tests. For the null model > section, the code is: > > obj <-SKAT_Null_Model(phenotype ~ covariates_matrix, out_type="D") > > where phenotype is a numeric binary of 0/1 for case/control status. Does it > matter if 0 is case or control and vice versa? > > Secondly, how exactly do you get the covariates matrix? For gene based > tests, what are the covariates? If all I have is patient ID, case/control > status and genotype data (0/1/2 or NA), how should I set this up? > Since SKAT is not a Bioconductor package, it's probably best to ask its maintainer, listed here: packageDescription("SKAT")[['Maintainer']] or here: http://cran.r-project.org/web/packages/SKAT/index.html Dan > Thanks, > > Cian > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor