docs/Setup.md

In hakyimlab/OmicKriging: Poly-Omic Prediction of Complex TRaits

Depedencies

Ensure the following R packages are installed on your environment. OmicKriging utilises these packages.

• ROCR
• doParallel
• parallel
• foreach
• iterators

NB: These libraries are required by the OmicKriging.

Phenotype file

The phenotype file used in your analysis should; - Contain phenotypes on the columns and samples on the rows - Not contain NA values, these break the analysis pipeline. - Mutate the NA values with column averages for each phenotype. - The individual IDs used in the phenotype file should be same as those in the grm file. - Contain an equal number or a subset of the individuals in the grm matrix. - E.g if a grm file has 997 individuals the phenotype should have 997 individuals or less. - If the phenotype file contains many individuals than those in the grm it will break the analysis pipeline.

• The sample ids column should be denoted as IID.
• The phenotype file should contain the sample id IID and family id FID in the first two columns. Phenotypes should start from column 3 to the end.

hakyimlab/OmicKriging documentation built on Nov. 1, 2020, 1:33 a.m.