From: Autism genetics: searching for specificity and convergence
References | Participants | De novovariant characteristics | Single genes implicated | Novel findings | Trends |
---|---|---|---|---|---|
Neale et al. 2012 [24] | 175 families*, simplex/multiplex and trios. 175 cases, 350 parents | Equal frequency of de novo mutations in cases and control | KATNAL2 and CHD8 harbor de novo mutations in cases, none found in controls (935 cases versus 870 controls; targeted exome sequencing) | Significantly enriched number of protein interactions among genes with missense or nonsense de novo mutations | Greater paternal and maternal age correlates with greater number of de novo mutations |
O'Roak et al. 2012 [22] | 209 families (SSC), simplex, trios and quads. 209 cases, 418 parents, 50 unaffected sibs | Equal frequency of de novo mutation in cases and controls | ASD cases harbor protein-disrupting mutations in GRIN2B, LAMC3, and SCN1A (mutation screening; 1,703 ASD cases, 744 controls) and CHD8 and NTNG1 (recurrent) | Genes with de novo mutations that cause missense or nonsense mutations form a β-catenin/chromatin remodeling protein network enriched for ASD candidate genes | 4:1 paternal origin of de novo mutations. Greater paternal age correlates with greater number of de novo SNVs |
Sanders et al. 2012 [25] | 238 families (SSC), simplex, trios and quads. 238 cases, 476 parents, 200 unaffected sibs | Equal frequency of de novo mutation in cases and controls | SCN2A significantly associated with ASD. KATNAL2, CHD8, and SCN2A significantly associated with ASD when combined with [22, 24] | Significantly greater non-synonymous and nonsense de novo SNVs in cases than unaffected sibs (all genes and brain-expressed genes), OR 1.93 for non-synonymous to silent SNVs in cases versus unaffected sibs | Greater paternal age correlates with greater number of de novo SNVs |
Iossifov et al. 2012 [23] | 343 families (SSC), simplex, quads. 343 cases, 686 parents, 343 unaffected sibs | Equal frequency of de novo mutation in cases and controls | KATNAL2, CHD8, SCN2A, DYRK1A, and POGZ significantly associated with ASD when combining all studies [22–25] | Twofold higher numbers of frame-shift, splice-site, and nonsense de novo mutations in cases than in unaffected sibs. Enriched number of gene-disrupting mutations in FMRP-associated genes ([23] alone and when combining data from all studies [22–25]) | Greater paternal and maternal age correlates with greater number of de novo mutations |