It has previously been shown in many clinical collections of patients with moderate to severe learning disabilities that rare CNVs can explain a proportion of these patients, typically 10-15%, depending on the phenotypic severity, presence of additional clinical features and resolution of the CNV discovery technology applied.

It remains an open question to what degree rare CNVs contribute to more mild learning difficulties. Learning difficulties can be variously defined, depending on the context (educational, clinical, social), but can include many problems affecting schoolwork, which may include specific learning difficulties such as dyslexia, dyscalculia or dyspraxia as well as more severe cognitive impairments.

Rare CNVs can be mined from primary intensity data from SNP GWAS chips. We propose to call rare CNVs from the existing ALSPAC genotype data held at the Sanger Institute, using a combination of algorithms to improve specificity.

We will then investigate the association of rare CNVs, both in terms of the presence/absence of large rare CNVs, and the overall genome-wide rare CNV burden, on various measures of cognitive performance reflecting learning difficulties.

It would be useful to know from the ALSPAC team if there are important covariates to account for when investigating these phenotypes.