Travis-CI Build Status GitHub license GitHub issues CRAN_Status_Badge CRAN_Logs_Rstudio

getmstatistic

M - An aggregate statistic, to identify systematic heterogeneity patterns and their direction of effect in meta-analysis

Overview

M quantitatively describes systematic (non-random) heterogeneity patterns acting across multiple variants in a GWAS meta-analysis. It’s primary use is to identify outlier studies, which either show “null” effects or consistently show stronger or weaker genetic effects than average across, the panel of variants examined in a meta-analysis.

M differs from conventional heterogeneity metrics (Q-statistic, I2), in that, it measures heterogeneity across multiple independently associated variants, whilst (Q-statistic, I2) measure heterogeneity at individual variants. Essentially, M measures systematic heterogeneity, whilst (Q-statistic, I2) measure random variant-specific heterogeneity.

Sources of systematic heterogeneity

Systematic heterogeneity can arise in a meta-analysis due to differences in the study characteristics of participating studies. Some of the differences may include: ancestry, allele frequencies, phenotype definition, age-of-disease onset, family-history, gender, linkage disequilibrium and quality control thresholds.

Practical benefits of exploring systematic heterogeneity

Installation: Stata

  1. Start stata
  2. To install the getmstatistic stata module: ssc install getmstatistic
  3. To get the example dataset in your current working directory: net get getmstatistic
  4. You’re all set, getmstatistic is installed, try some of the examples in the getmstatistic help file

Tip! getmstatistic depends on the following user-written Stata commands which can be installed using ssc install package-name or findit package-name:

A full list of getmstatistic dependencies can be found in the help file.


1. Unzip the folder

Tip! the folder should contain the following files: getmstatistic.ado getmstatistic.sthlp heartgenes214.dta

2. Start stata

3. Locate your personal directory where stata stores user generated files by typing: `sysdir`
sysdir

Tip! on mac the ado/personal directory is likely to be at: ~/Library/Application Support/Stata/ado/personal/
for linux: ~/ado/personal/ 
for windows: c:\ado\personal\

4. Copy getmstatistic.ado and getmstatistic.sthlp to the g sub-directory in personal

Tip! if the g sub-directory does not exist, that just means you do not have user generated commands
that start with the letter g. In that case create a folder named g in the personal directory.

5. Type help getmstatistic in Stata to open the getmstatistic help file.

6. Load the example dataset: heartgenes214.dta

7. You're all set, getmstatistic is installed, try some of the examples in the getmstatistic help file

Installation: getmstatistic R package

# To install the release version from CRAN:
install.packages("getmstatistic")

# Load libraries
library(getmstatistic)  # for calculating M statistics
library(gridExtra)      # for generating tables


# To install the development version from GitHub:

# install devtools
install.packages("devtools")

# install getmstatistic
library(devtools)
devtools::install_github("magosil86/getmstatistic")

# Load libraries
library(getmstatistic)  # for calculating M statistics
library(gridExtra)      # for generating tables

Usage

Details

Magosi LE, Goel A, Hopewell JC, Farrall M, on behalf of the CARDIoGRAMplusC4D Consortium (2017) Identifying systematic heterogeneity patterns in genetic association meta-analysis studies. PLoS Genet 13(5): e1006755. https://doi.org/10.1371/journal.pgen.1006755.

Getting help

To suggest new features, learn about getmstatistic updates, report bugs, ask questions about the mstatistic, or just interact with other users, sign up to the getmstatistic mailing list.

Code of conduct

Contributions are welcome. Please observe the Contributor Code of Conduct when participating in this project.

Citation

Magosi LE, Goel A, Hopewell JC, Farrall M, on behalf of the CARDIoGRAMplusC4D Consortium (2017) Identifying systematic heterogeneity patterns in genetic association meta-analysis studies. PLoS Genet 13(5): e1006755. https://doi.org/10.1371/journal.pgen.1006755.

Acknowledgements.

Roger M. Harbord’s metareg command for computation of standardized predicted random effects which are then incorporated into calculations for the M statistics. Harbord, R. M., & Higgins, J. P. T. (2008). Meta-regression in Stata. Stata Journal 8: 493‚Äì519.

Authors.

Lerato E. Magosi, Jemma C. Hopewell and Martin Farrall.

Maintainer.

Lerato E. Magosi lmagosi@well.ox.ac.uk or magosil86@gmail.com

License

See the LICENSE file.