Subprograms

BridgePRS consists of five related multi-function subprograms:

1) BridgePRS prs-single
2) BridgePRS build-model
3) BridgePRS prs-port
4) BridgePRS prs-prior
5) BridgePRS analyze combine

Subprogram	Input	Subcommands	Output
prs-single	Target Pop Data	run,clump,beta,predict,quantify	prs-result
build-model	Model Pop Data	**run**clump,beta,predict,prior	model-params
prs-port	Target Pop + Model Result	run,predict,quantify	prs-port-result
prs-prior	Target Pop Data + Model Result	run,clump,beta,test,predict	prs-prior-result
analyze	PRS Result Files	**run**result,combine	single-plot,weighted prs-result,weighted prs plot

These programs can be called consecutively using the easyrun master-program which will create the following directory structure:

prs-single

The subprogram (prs-single) and subcommand (run) (./bridgePRS prs-single run) requires that the following target population data be provided on the command line or inside of a configuration file:

1. --pop: The name of your target population
2. --ldpop: The ld reference name, if different from target population name
3. --sumstats_prefix: Sumstats data
4. --genotype_prefix: Target Genotype Data
5. --phenotype_file: Target Phenotype File

This subprogram performs single population RidgePRS.

build-model

The subprogram is run on the base population (where the GWAS is larger).

1. --pop: The name of your target population
2. --ldpop: The ld reference name, if different from target population name
3. --sumstats_prefix: Base sumstats data
4. --genotype_prefix: Base Genotype Data (If available)
5. --phenotype_file: Base Phenotype File (If available)

This program performs RidgePRS on the base population and then provides a prior distribution for each SNP weight that can be passed onto the next programs.

prs-port

The subprogram uses the model result and the target population data to "port" the SNP weights from the base population to make estimates in the target population.

prs-prior

The subprogram uses the model result and the target population data and uses the the SNP weights from the base population to estimate a prior distribution to make estimates in the target population.

analyze

The analyze subprogram is used to either make a plot from a single result (above) or to combine the results of all three.