Empowering multitrait brain phenotypes GWAS
Auvergne, A.; Traut, N.; Julienne, H.; Frouin, A.; Toro, R.; Aschard, h.
Show abstract
Brain magnetic resonance imaging (MRI) data are now becoming available in human genetic cohorts including thousands of participants, providing a powerful mean to decipher the genetic architecture of brain phenotypes. Univariates genome-wide association screenings (GWAS) are typically applied as a baseline approach, but there are increasing interest for multitrait approaches to both increase statistical power and uncover shared structure underlying those phenotypes. However, there are limited guidelines on how to select phenotypes to be analyzed jointly and optimize power. Here we investigated factors impacting variant discoverability of state-of-the art multitrait GWAS using 1,010 MRI-derived phenotypes measured in 32,947 participants from the UK Biobank. Our study shows that the expected gain of multitrait approaches over univariate GWAS is highly predictable. We further propose data-driven strategy to define optimize sets and compare its performances with biologically driven approaches.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Uncovering causal gene-tissue pairs and variants: A multivariable TWAS method controlling for infinitesimal effects 96%
- Testing and controlling for horizontal pleiotropy with the probabilistic Mendelian randomization in transcriptome-wide association studies 96%
- Identification of putative causal loci in whole-genome sequencing data via knockoff statistics 96%
Similar papers in this journal
- Controlling for background genetic effects using polygenic scores improves the power of genome-wide association studies 96%
- Qtlizer: comprehensive QTL annotation of GWAS results 94%
- Genome-wide association studies of polygenic risk score-derived phenotypes may lead to inflated false positive rates 94%
Similar papers in this journal
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.