Gene based P-value Adaptive Combination Approach (GPA)

GPA can integrate association evidence from summary statistics, either p value or other statistical value, continuous or binary traits, which might come from the same or different studies of traits. To run GPA, you need summary statistic data in a file. Then issue a command such as:

gpa_openmp -i test.txt -o pvalue.txt -p 10000
gpa_mpi -i test.txt -o pvalue.txt -p 10000

Current Version

• v1.0

Setup and Installation

Basic Dependencies

1. GNU GCC or Intel C++ with C++11 support i.e. gcc4.9+ or above, icpc15.0+ or above.
2. MPI Library with MPI-3 support i.e. OpenMPI 1.8 and above or cray-mpich/7.4.0 and above. By default the mpic++ wrapper is needed. If you are on a Cray cluster and the wrapper is "CC". You will need to edit the compiler.mk file. Uncomment the line "MCC := CC" and comment out "MCC := mpic++".

Installation Steps

1. Download the tarball with compiled executables for Linux with GCC 4.9 and above from https://github.com/Biocomputing-Research-Group/GPA/releases. The code has been tested only on Linux.
2. If you decide to download the source code, use the following commands to build:
3. OpenMP version "make openmp".
4. MPI version version "make mpi"
5. All the versions can be built with "make all" If compiled successfully, the required executables will be in bin directory.

Prepare summary statistics

A toy example summary statistic data test.txt is provided in "toy_example". The format of this input data is as follows:

1st line, gene name

2nd line, Z-statistic

3rd line, P-value

4th line, covariance matrix

Running GPA

There are two basic versions of GPA: one for running on a single machine and another for running with MPI on a cluster.

• Single Machine Version: This version of GPA should be used if you are going to run on a single machine with one or more cores. The quick start command as shown below will be typed on the command line terminal.

#!/bin/bash

gpa_openmp -i test.txt -o pvalue.txt -p 10000

Results ('pvalue.txt' files) will be saved on the output directory. Use gpa_openmp -h for help information.

• MPI Version: This version of GPA should be used if you are going to run on a cluster with MPI support. An example bash script submit_job.pbs is provide in "toy_example".

The quick start commands are:

### MPI Verion 
gpa_mpi -i test.txt -o pvalue.txt -p 10000

Results ('pvalue.txt' files) will be saved on the output directory. Use gpa_mpi -h for help information.

Miscellany

Questions?

• Xuan Guo