Human Genetics

, Volume 129, Issue 1, pp 59–70

Genome-scan for IQ discrepancy in autism: evidence for loci on chromosomes 10 and 16

Authors

  • Nicola H. Chapman
    • Department of MedicineUniversity of Washington
  • Annette Estes
    • Department of Psychiatry and Behavioral SciencesUniversity of Washington
  • Jeff Munson
    • Department of Psychiatry and Behavioral SciencesUniversity of Washington
  • Raphael Bernier
    • Department of Psychiatry and Behavioral SciencesUniversity of Washington
  • Sara J. Webb
    • Department of Psychiatry and Behavioral SciencesUniversity of Washington
  • Joseph H. Rothstein
    • Department of MedicineUniversity of Washington
  • Nancy J. Minshew
    • Department of PsychiatryUniversity of Pittsburgh School of Medicine
    • Department of NeurologyUniversity of Pittsburgh School of Medicine
  • Geraldine Dawson
    • Autism SpeaksUniversity of North Carolina Chapel Hill
    • Department of PsychiatryUniversity of North Carolina Chapel Hill
  • Gerard D. Schellenberg
    • Department of Pathology and Laboratory MedicineUniversity of Pennsylvania School of Medicine
    • Department of MedicineUniversity of Washington
    • Department of BiostatisticsUniversity of Washington
    • Department of Genome SciencesUniversity of Washington
Original Investigation

DOI: 10.1007/s00439-010-0899-z

Cite this article as:
Chapman, N.H., Estes, A., Munson, J. et al. Hum Genet (2011) 129: 59. doi:10.1007/s00439-010-0899-z

Abstract

Performance IQ (PIQ) greater than verbal IQ (VIQ) is often observed in studies of the cognitive abilities of autistic individuals. This characteristic is correlated with social and communication impairments, key parts of the autism diagnosis. We present the first genetic analyses of IQ discrepancy (PIQ–VIQ) as an autism-related phenotype. We performed genome-wide joint linkage and segregation analyses on 287 multiplex families, using a Markov chain Monte Carlo approach. Genetic data included a genome-scan of 387 micro-satellite markers in 210 families augmented with additional markers added in a subset of families. Empirical P values were calculated for five interesting regions. Linkage analysis identified five chromosomal regions with substantial regional evidence of linkage; 10p12 [P = 0.001; genome-wide (gw) P = 0.05], 16q23 (P = 0.015; gw P = 0.53), 2p21 (P = 0.03, gw P = 0.78), 6q25 (P = 0.047, gw P = 0.91) and 15q23-25 (P = 0.053, gw P = 0.93). The location of the chromosome 10 linkage signal coincides with a region noted in a much earlier genome-scan for autism, and the chromosome 16 signal coincides exactly with a linkage signal for non-word repetition in specific language impairment. This study provides strong evidence for a QTL influencing IQ discrepancy in families with autistic individuals on chromosome 10, and suggestive evidence for a QTL on chromosome 16. The location of the chromosome 16 signal suggests a candidate gene, CDH13, a T-cadherin expressed in the brain, which has been implicated in previous SNP studies of autism and ADHD.

Supplementary material

439_2010_899_MOESM1_ESM.pdf (14 kb)
Online Resource 1: Plots of ADOS and ADI communication domain and social domain scores as a function of IQ discrepancy (PDF 14 kb)
439_2010_899_MOESM2_ESM.doc (36 kb)
Online Resource 2: Table of single-marker linkage results in regions of interest (DOC 36 kb)

Copyright information

© Springer-Verlag 2010