No evidence for linkage of liability to autism to HOXA1 in a sample from the CPEA network

A recent study by Ingram et al. [2000b: Teratology 62:393-405] suggests a (His)73(Arg) polymorphism (A:G) in HOXA1 contributes substantially to a liability for autism. Using 68 individuals diagnosed with Autism Spectrum Disorders, they found a significant dearth of G homozygotes and biased transmission of G alleles from parents to affected offspring, especially from mothers. Because the connection between HOXA1 and liability to autism is compelling, we attempted to replicate their finding using a larger, independent sample from the Collaborative Programs of Excellence in Autism (CPEA) network. In our data, genotype frequencies conform to Hardy-Weinberg equilibrium; allele transmissions meet Mendelian expectations; and there is no obvious sex-biased allele transmission. Based on our sample size, calculations suggest that we would have at least 95% power to detect linkage and association even if the A:G polymorphism were to account for only 1% of the heritability of autism. Therefore, although we cannot exclude the possibility that the samples in the two studies are intrinsically different, our data from our sample argue against a major role for HOXA1 (His)73(Arg) in liability to autism.