This package provides the framework used within ALSPAC/ARIES for running meQTL analysis
Python 2.7+ (do not use Python 3)
PLINK v1.9 needs to be on your path
The following R Packages are also required - these can be installed using the install.packages command during an interactive R session:
- argparse
- foreach
- doMC
- parallel
- MatrixEQTL
To use the package, just type
python run.py
You can see the list of accepted arguments using the --help parameter, i.e.
python run.py --help
--geno: path to the binary ped file, i.e. PLINK "bfile"
--meth: path to the mathylation data in Rdata format
--covars: path to the covariates used to residualize the methylation data.
Note: covariate data must be a tab-delimited file with samples as rows and covariates as columns. Also, "Sex" and "Batch" MUST be covariates (proprocessing.R expects them and will fail if they are omitted)
--tp: a file listing the sample IDs in the ARIES time point to analyse
--fo: where the output is written
--keep-snps: path to a file containing the list of SNPs to analyze (defaults to all SNPs)
--keep-probes: path to a file containing a list of CpG probes to analyze (defaults to all probes)
--pv: the p-value threshold used when reporting results (defaults to 1e-5)
--distance: the distance used when flagging associations as trans (defaults to 1Mb)
These arguments do not affect the results, but do provide developers with some debugging options:
--batch: the batch size to use (defaults to 10,000)
--skip-annotation: skip the final annotation step to save some time
--keep-intermediate-data: do not delete intermediate data files when finished
- run meQTL analysis in 15up for all SNPs/probe combination:
python run.py --geno <path_to_geno> --meth <path_to_meth> --covars <path_to_covars> --tp 15up --fo <path_to_output>
- run meQTL analysis in FOM for all SNPs, but a handful of probes:
python run.py --geno <path_to_geno> --meth <path_to_meth> --covars <path_to_covars> --tp FOM --fo <path_to_output> --keep-probes <path_to_probes>
- run meQTL analysis in F7 for all probes, but a handful of SNPs:
python run.py --geno <path_to_geno> --meth <path_to_meth> --covars <path_to_covars> --tp F7 --fo <path_to_output> --keep-snps <path_to_snps>