Hypothesis-driven genome-wide association studies provide novel insights into genetics of reading disabilities

Kaitlyn M Price,Karen G Wigg,Else Eising,Yu Feng,Kirsten Blokland,Margaret Wilkinson,Elizabeth N Kerr,Sharon L Guger,Quantitative Trait Working Group of the GenLang Consortium,Simon E Fisher,Maureen W Lovett,Lisa J Strug,Cathy L Barr,Filippo Abbondanza,Andrea G Allegrini,Till F M Andlauer,Timothy C Bates,Manon Bernard,Milene Bonte,Dorret I Boomsma,Thomas Bourgeron,Daniel Brandeis,Manuel Carreiras,Fabiola Ceroni,Valéria Csépe,Philip S Dale,John C DeFries,Peter F de Jong,Jean Francois Démonet,Eveline L de Zeeuw,Marie-Christine J Franken,Clyde Francks,Margot Gerritse,Alessandro Gialluisi,Scott D Gordon,Jeffrey R Gruen,Marianna E Hayiou-Thomas,Juan Hernández-Cabrera,Jouke-Jan Hottenga,Charles Hulme,Philip R Jansen,Juha Kere,Tanner Koomar,Karin Landerl,Gabriel T Leonard,Zhijie Liao,Michelle Luciano,Heikki Lyytinen,Nicholas G Martin,Angela Martinelli,Urs Maurer,Jacob J Michaelson,Nazanin Mirza-Schreiber,Kristina Moll,Anthony P Monaco,Angela T Morgan,Bertram Müller-Myhsok,Dianne F Newbury,Markus M Nöthen,Richard K Olson,Silvia Paracchini,Tomas Paus,Zdenka Pausova,Craig E Pennell,Bruce F Pennington,Robert J Plomin,Franck Ramus,Sheena Reilly,Louis Richer,Kaili Rimfeld,Gerd Schulte-Körne,Chin Yang Shapland,Nuala H Simpson,Shelley D Smith,Margaret J Snowling,Beate St Pourcain,John F Stein,Joel B Talcott,Henning Tiemeier,J Bruce Tomblin,Dongnhu T Truong,Elsje van Bergen,Marc P van der Schroeff,Marjolein Van Donkelaar,Ellen Verhoef,Carol A Wang,Kate E Watkins,Andrew J O Whitehouse,Erik G Willcutt,Margaret J Wright,Gu Zhu

DOI: https://doi.org/10.1038/s41398-022-02250-z

2022-11-29

Abstract:Reading Disability (RD) is often characterized by difficulties in the phonology of the language. While the molecular mechanisms underlying it are largely undetermined, loci are being revealed by genome-wide association studies (GWAS). In a previous GWAS for word reading (Price, 2020), we observed that top single-nucleotide polymorphisms (SNPs) were located near to or in genes involved in neuronal migration/axon guidance (NM/AG) or loci implicated in autism spectrum disorder (ASD). A prominent theory of RD etiology posits that it involves disturbed neuronal migration, while potential links between RD-ASD have not been extensively investigated. To improve power to identify associated loci, we up-weighted variants involved in NM/AG or ASD, separately, and performed a new Hypothesis-Driven (HD)-GWAS. The approach was applied to a Toronto RD sample and a meta-analysis of the GenLang Consortium. For the Toronto sample (n = 624), no SNPs reached significance; however, by gene-set analysis, the joint contribution of ASD-related genes passed the threshold (p~1.45 × 10-2, threshold = 2.5 × 10-2). For the GenLang Cohort (n = 26,558), SNPs in DOCK7 and CDH4 showed significant association for the NM/AG hypothesis (sFDR q = 1.02 × 10-2). To make the GenLang dataset more similar to Toronto, we repeated the analysis restricting to samples selected for reading/language deficits (n = 4152). In this GenLang selected subset, we found significant association for a locus intergenic between BTG3-C21orf91 for both hypotheses (sFDR q < 9.00 × 10-4). This study contributes candidate loci to the genetics of word reading. Data also suggest that, although different variants may be involved, alleles implicated in ASD risk may be found in the same genes as those implicated in word reading. This finding is limited to the Toronto sample suggesting that ascertainment influences genetic associations.