Depends:
The goal of this program is to take a genome (simulated using HapGen2
off of a 1KG reference genome), superimpose an imaginary genetic condition, blind ourselves to the basis of this condition and attempt to identify the contributing variants through association analysis. We can then use different multiple comparrisons techniques and attempt to identify
- Methods which identify more true positives while maintaining FDR or FWER
- Contributing covariates to the disovery of variants
Full details of the implimentation may be found in the associated preprint -- please contact the maintainors for a current version.
The first priority is to understand how to take input arguements and parse them. We have a lot of parameters that need to be incorporated unfortuantely.
An example of intended usage:
./mchg --gen $GEN_FILE \
--summary $SUMMARY_FILE \
--output $OUTPUT_STREAM \
--h2 $HERITABILITY \
--pc $PERCENT_CAUSAL \
--pnc $PERCENT_NON_CAUSAL \
--nc $NUMBER_CAUSAL \
--maf $MAF \
--maf-range $MAF_RANGE \
--nrep $NREP
Variables:
$GEN_FILE
: File path to the.gen
file.- File reference
std::string
$SUMMARY_FILE
: File path to the.summary
file.- Output of
snptest --summary-stats-only
, reference here std::string
- Output of
$OUTPUT_STREAM
: File path to the output stream for depositing summary statistics.std::string
$HERITABILITY
: The heritability of the simulated condition.double
in [0,1]
$PERCENT_CAUSAL
: Percent of the causal variants to be included in the second strata.double
in [0,1]
$PERCENT_NON_CAUSAL
: Percent of the non-causal variants to be included in the second strata.double
in [0,1]- I know these two are confusing, but couldn't think of a better way to do this.
$NUMBER_CAUSAL
: Number of causal SNPs to be simulatedint
$MAF
: Simulate based on MAF bins or notbool
$MAF_RANGE
: Vector of length two; lower end of the MAF and upper end of the MAF bin to be generated.arma::vec
$NREP
: Number of permutations to performint
Secondly