Causal effect estimation from data simulating instrument correlations, instrument strength, and horizontal pleiotropy.

Causal effect estimation from data simulating instrument correlations, instrument strength, and horizontal pleiotropy.

<p><b>(A)</b> Causal diagram for the simulation of two exposures <i>X</i><sub>1</sub> and <i>X</i><sub>2</sub> for an outcome <i>Y</i>, with two shared instruments <i>E</i><sub>1</sub> and <i>...

Full description

Saved in:

Bibliographic Details
Main Author:	Mariyam Khan (20161530) (author)
Other Authors:	Adriaan-Alexander Ludl (9321107) (author), Sean Bankier (15152851) (author), Johan L. M. Björkegren (7759127) (author), Tom Michoel (29810) (author)
Published:	2024
Subjects:	Medicine Genetics Molecular Biology Virology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified variance small even structural equation models multivariate mendelian randomization linkage disequilibrium ), coronary artery disease individual causal effects direct causal effects confirm causal roles pleiotropic regulatory effects extensive simulations confirmed predicting causal genes wide significant loci disease risk loci multiple nearby genes multiple colocated genes multiple exposures collectively locus still require div >< p >- eqtl associations predict causal gene instrumental set condition correlated instruments satisfies considering risk loci causal genes multiple exposures causal variants multiple tissues candidate genes regulatory pleiotropy instrumental variables highly correlated extensive degree twelve loci gwas loci genomic loci within tissues theoretical results statistical technique starnet study standard mr single model seven vascular phactr1 </ metabolic tissues limited number horizontal pleiotropy given locus genetic variants genetic instruments field dictates classical result cis </ bias introduced arterial tissues among others adamts7 </
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	<p><b>(A)</b> Causal diagram for the simulation of two exposures <i>X</i><sub>1</sub> and <i>X</i><sub>2</sub> for an outcome <i>Y</i>, with two shared instruments <i>E</i><sub>1</sub> and <i>E</i><sub>2</sub> with pairwise covariance <i>α</i> and matrix of instrument effect sizes <i>A</i> = [<i>a</i><sub><i>ij</i></sub>]. <b>(B)</b> Difference () between estimated and true causal effect of <i>X</i><sub>1</sub> (true effect size <i>c</i> = 0.2) on <i>Y</i> across 1,000 independently simulated datasets for a range of sample sizes. <b>(C,D)</b> Distribution of the difference between estimated and true causal effects for <i>X</i><sub>1</sub> (left; true effect size <i>c</i> = 0.2) and <i>X</i><sub>2</sub> (right; true effect size <i>c</i> = 0.6) across 1,000 independently simulated datasets and a range of sample sizes comparing strong (0.1 − 0.3) vs weak (0.001 − 0.01) instruments. <b>(E)</b> False causal diagram used for inference where one causal exposure is missing. <b>(F-G)</b> Difference between estimated and true causal effect of <i>LPA</i> (left, true effect size <i>c</i><sub><i>LPA</i></sub> = −0.05) and <i>PLG</i> (right, true effect size <i>c</i><sub><i>PLG</i></sub> = −0.27) on <i>Y</i> across 1,000 independently simulated datasets with sample size 2000 across a range of effect sizes of the hidden gene <i>SLC22A3</i> mediating horizontal pleiotropic effects of the instruments on <i>Y</i>, showing both the estimates with the correct model (green) and the estimates with the hidden exposure model (orange). See <a href="http://www.plosgenetics.org/article/info:doi/10.1371/journal.pgen.1011473#sec002" target="_blank">Methods</a> for simulation details.</p>

Cannot write session to /tmp/vufind_sessions/sess_842k1i5f3fr95c75n8402a20bm