Metasubtract: an R‐package to analytically produce leave‐one‐out meta‐analysis GWAS summary statistics

Abstract Summary Summary statistics from a meta‐analysis of genome‐wide association studies (meta-GWAS) can be used for many follow-up analyses. One valuable application is the creation of polygenic scores. However, if polygenic scores are calculated in a validation cohort that was part of the meta-...

Full description

Saved in:
Bibliographic Details
Published in:Bioinformatics Vol. 36; no. 16; pp. 4521 - 4522
Main Author: Nolte, Ilja M
Format: Journal Article
Language:English
Published: England Oxford University Press 15.08.2020
Subjects:
Applications Notes
Genome-Wide Association Study
Genotype
Humans
Multifactorial Inheritance
Polymorphism, Single Nucleotide
Sample Size
Software
ISSN:1367-4803, 1367-4811, 1460-2059, 1367-4811
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract Summary Summary statistics from a meta‐analysis of genome‐wide association studies (meta-GWAS) can be used for many follow-up analyses. One valuable application is the creation of polygenic scores. However, if polygenic scores are calculated in a validation cohort that was part of the meta-GWAS consortium, this cohort is not independent and analyses will therefore yield inflated results. The R package ‘MetaSubtract’ was developed to subtract the results of the validation cohort from meta‐GWAS summary statistics analytically. The statistical formulas for a meta‐analysis were inverted to compute corrected summary statistics of a meta‐GWAS leaving one (or more) cohort(s) out. These formulas have been implemented in MetaSubtract for different meta‐analyses methods (fixed effects inverse variance or square root sample size weighted z‐score) accounting for no, single or double genomic control correction. Results obtained by MetaSubtract correlate very well to those calculated using the traditional way, i.e. by performing a meta‐analysis leaving out the validation cohort. In conclusion, MetaSubtract allows researchers to compute meta‐GWAS summary statistics that are independent of the GWAS results of the validation cohort without requiring access to the cohort level GWAS results of the corresponding meta-GWAS consortium. Availability and implementation https://cran.r-project.org/web/packages/MetaSubtract. Supplementary information Supplementary data are available at Bioinformatics online.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:1367-4803
1367-4811
1460-2059
1367-4811
DOI:10.1093/bioinformatics/btaa570