“Estimating the Effect-Size of Gene Dosage on Cognitive Ability across the Coding Genome”, Guillaume Huguet, Catherine Schramm, Elise Douard, Tamer Petra, Antoine Main, Pauline Monin, Jade England, Khadije Jizi, Thomas Renne, Myriam Poirier, Sabrina Nowak, Charles-Olivier Martin, Nadine Younis, Inga Sophia Knoth, Martineau Jean-Louis, Zohra Saci, Maude Auger, Frédérique Tihy, Géraldine Mathonnet, Catalina Maftei, France Léveillé, David J. Porteous, Gail Davies, Paul Redmond, Sarah E. Harris, W. David Hill, Emmanuelle Lemyre, Gunter Schumann, Thomas Bourgeron, Zdenka Pausova, Tomas Paus, Sherif Karama, Sarah Lippe, Ian J. Deary, Laura Almasy, Aurélie Labbe, David Glahn, Celia M. T. Greenwood, Sébastien Jacquemont2020-04-05 (, , ; similar)⁠:

Rare genomic Copy Number Variants (CNVs) are major contributors to neurodevelopmental disorder. The vast majority of pathogenic CNVs reported back to patients are ultra-rare and their quantitative effects on traits such as intelligence are undocumented.

Here, we identified all CNVs ≥ 50 kilobase in 24,092 individuals from unselected and autism cohorts. We developed statistical models to estimate the effect-size of CNVs on intelligence based on their coding and non-coding characteristics.

Measures of intolerance to haploinsufficiency best explained the effect of any deletion or duplication on general intelligence. There was no heterogeneity across unselected and autism cohorts. Validation was performed using an intraclass concordance and showed that model estimates of general intelligence were 78% accurate with mean effect-sizes previously published for 47 CNVs.

Inheritance data on 27,766 CNVs showed that deletions and duplications with the same large effect-size on intelligence occur de novo at the same frequency.

Our first outline for the effect sizes of all coding genes on intelligence suggests that around 10,000 genes affect this trait.