Epistasis Pipeline

Running the Epistasis Pipeline on UW-Madison Clusters

1. add files (tfam, tped, pheno, covar) to the epistasis/data/ directory
2. copy fixpheno.sh file into the epistasis/data/ directory
3. run the command sh fixpheno.sh PREFIX where PREFIX is the actual prefix for the data set
4. move to the epistasis/ directory
5. run the command python epistasis_pipeline.py PREFIX --covar (with additional arguments as necessary)
   Note: for a detailed list of additional arguments, run the command python epistasis_pipeline.py -h
6. wait while pipeline runs
7. transfer the directory results/ back to the server
8. check the condor_out/ directory for any errors
9. clear out the data/ directory for next time

NOTE:

• Sometimes the "condor_q" command fails and quits the program unexpectedly. In this case, rerun the pipeline
• Ensure that files scripts/fastlmmc, scripts/plink, fixpheno.sh are executable (green). If not run the command chmod +x FILE_NAME