biobank-plink-assoc

This repository provides a tool to perform Genome-wide-association study (GWAS). The data analysis is achived using plink. The contribution of this repository is that

1. convert\(ph\g)enotype_convert: A more flexible parsing tool that can read formatted text outputs with less stringent format requirment. For example, Data from UK Biobank can be parsed with the right parameters.
2. convert-assoc-qt: A rudimentary cross-platform GUI using Qt that guides the user through the steps of performing association studies.

Components and their relations

├───CMakeLists.txt: Cmake for Qt GUI program
├───convert: library for parsing texts to plink-compatible format
│   ├───genotype_convert.h/.cpp
│   └───phenotype_convert.h/.cpp
├───convert-assoc-qt: GUI program
├───plink_1.9:
│   └───CMakeLists.txt: "Binary dependency-free" version of plink
├───sample_input: Sample input files to test functionality
├───test-vsproj: tests for convert (Visual Studio project only)
└───zlib-1.2.11: Dependency for plink
    └───CMakeLists.txt

Installation guide

As of now, we do not provide an installer or a binary build, and Qt is required to build the program. Install Qt to get started.

On Windows machines, use MinGW or LLVM toolkits to build this program, as MSVC cannot compile plink.

Note: The main program uses plink as a binary executable; however, build convert_assoc project does not build plink automatically. Either build all projects at least once to build it, or provide an alternative plink binary if you wish.

Usage guide

The program is (hopefully) self-explanatory. The program guides the user through the flowchart to

1. Parse the genotype and phenotype files with varying column layouts and delimiters
2. Choose a binary partition of phenotypes in the case of discrete phenotypes

A video demo using the sample inputs is in the plan.