Reference
"Genome-wide association analysis of diverticular disease points towards neuromuscular,
connective tissue and epithelial pathomechanisms"
Clemens Schafmayer, James W. Harrison, Stephan Buch, Christina Lange, Matthias C. Reichert,
Philipp Hofer, François Cossais, Juozas Kupcinskas, Witigo von Schönfels, Bodo Schniewind,
Wolfgang Kruis, Jürgen Tepel, Myrko Zobel, Jonas Rosendahl, Thorsten Jacobi, Andreas Walther-Berends,
Michael Schroeder, Ilka Vogel, Petr Sergeev, Hans Bödeker, Holger Hinrichsen, Andreas Volk,
Jens-Uwe Erk, Greta Burmeister, Alexander Hendricks, Sebastian Hinz, Sebastian Wolff,
Martina Böttner, Andrew R. Wood, Jessica Tyrrell, Robin N. Beaumont, Melanie Langheinrich,
Torsten Kucharzik, Stefanie Brezina, Ursula Huber-Schönauer, Leonora Pietsch, Laura Sophie Noack,
Mario Brosch, Alexander Herrmann, Raghavan Veera Thangapandi, H.-Wolfgang Schimming,
Sebastian Zeissig, Stefan Palm, Gerd Focke, Anna Andreasson, Peter Thelin Schmidt,
Jürgen Weitz, Michael Krawczak, Henry Völzke, Gernot Leeb, Patrick Michl, Wolfgang Lieb,
Robert Grützmann, Andre Franke, Frank Lammert, Thomas Becker, Limas Kupcinskas, Mauro D’Amato,
Thilo Wedel, Christian Datz, Andrea Gsur, Michael N Weedon, Jochen Hampe
Schafmayer C, Harrison JW, Buch S, et al. Gut 2019;0:1–12. doi:10.1136/gutjnl-2018-317619
Additional Supplementary Material for the manuscript
Primary Discovery GWAS analysis :
GWAS_summary_1-23.dosages.maf_0.01.info_0.4.txt.gz
Genome-wide association analysis of 31,917 cases with diverticular disease (ICD-10 K57.-) and 419,135 controls from the UK Biobank.
Header abbreviations are detailed below.
The analysis was conducted using BOLT-LMMv2.3 [1].
Table headers :
| SNP | rs number |
| CHR | chromosome |
| BP | physical (base pair) position (GRCh37) |
| ALLELE1 | effect allele |
| ALLELE0 | non-effect allele |
| A1FREQ | allele frequency of ALLELE1 |
| INFO | imputation quality score |
| BETA | effect size from BOLT-LMM approximation to infinitesimal mixed model (ALLELE1) |
| SE | standard error of effect size |
| P | P_BOLT_LMM: non-infinitesimal mixed model association test p-value |
Filter applied :
minor allele frequency (MAF) >1%
imputation quality score (INFO) >0.4
Estimation of odds ratios :
Because BOLT-LMM still uses a linear model (rather than a logistic model) when analyzing case-control traits, a transformation is required in order to convert SNP effect size estimates (“betas”) on the quantitative scale to traditional odds ratios. A reasonable approximation is:
log OR = β / (μ * (1 - μ)), where μ = case fraction.
cases: 31917
controls: 419135
Standard errors of SNP effect size estimates should also be divided by (μ * (1 - μ)) when applying the above transformation to obtain log odds ratios.
[1] Loh PR, et al. (2015). Efficient Bayesian mixed-model analysis increases association power in large cohorts. Nature Genetics. 47, 284-290 (2015).