R code for selected results from "Two-phase sample selection strategies for design and analysis in post-genome wide association fine-mapping studies" by Espin-Garcia, Craiu, and Bull (2021).
This repository contains two files and two folders:
- twoPhaseGAS_1.07.tar.gz
- example_Poisson.R
- Sim2_PracticalScenario
- App_NFBC66
This file contains the source code for a preliminary version of an R package implementing the proposed methods. This package is necessary for the codes below. A description on how to install the package is contained in file Sim2_PracticalScenario/Realistic_Simulation.R
This file contains an example on how to obtain the GA design under Poisson regression modelling for which no optimal design has been previously described.
Contains the necessary codes and information to reproduce the results in Section 4 of the manuscript. The main code is "Realistic_Simulation.R" while files "Step_[1-4].R" contain the auxiliary functions with the details of each step. These scripts rely on file "data_Realistic_R=1K_N=5K.RData", which containts the simulation replicates, to fully reproduce the results from the manuscript. This file is available as Supplementary Data along with the manuscript. However, the file can be generated using script "Step_0_Data_generation.R", although the results will not be reproducible. Files "genetic_map_GRCh37_chr16.txt.gz" and "EUR.chr16.HERPUD1-CETP.genotypes.vcf.gz" can be used to generate such new set of replicates.
This folder contains detailed bash scripts and R codes that describe the analysis steps for the illustration on the North Finland Birth Cohort of 1966 (NFBC66) data. Note that no data from the study were made available. Researchers can request access to these data through dbGAP on this link, which containts more information of the study.