Building PRS Models Based on Summary Statistics of GWAs

Shrinkage estimator for polygenic risk prediction (PRS) models based on summary statistics of genome-wide association (GWA) studies. Based upon the methods and original 'PANPRS' package as found in: Chen, Chatterjee, Landi, and Shi (2020) .

PANPRS

Installation

For the sparse matrix implementation, please install the R package as follows:

devtools::install_github("katherine-h-l/PANPRSnext@sparse", force = TRUE)

For the dense matrix implementation, please install the R package as follows:

devtools::install_github("katherine-h-l/PANPRSnext@master", force = TRUE)

Input for PANPRS incorporating multiple traits and functional annotations of SNPs.

summaryZ, The Z statistics of p SNPs from q GWA studies. A matrix with dimension p x q for p SNPs and q traits. The first column corresponds to the primary trait and the rest columns correspond to the secondary traits.

Nvec, A vector of length q for the sample sizes of q GWA studies.

plinkLD, LD matrix information.

NumIter, The number of maximum iterations for the estimation procedure.

funcIndex, Inputs for the functional annotations of SNPs. A p x k matrix with (0,1) entry; p is the number of SNPs and k is the number of functional annotations. For the element at i-th row, j-th column, the entry 0 means SNP i without j-th functional annotation; entry 1 means otherwise.

numfunc, The number of functional annotations.

dfMax The upper bound of the number of non-zero estimates of coefficients for the primary trait.

Usage:

The current version only work on Unix, Linux and Mac System, R(>=3.4.3), R packages "gtools" and "permutations" and GCC(>=4.4.7) are required.

Modify the parameters in the gsfPEN.R, and execute it.

Example

Please find it in the R package.