Objectives: to assess the frequency and genomic location of segmental CNVs in neocortical tissue samples from autistic patients and matched controls, and to assess its degree of correlation with gene expression.

Methods: CNV analysis was performed on genomic DNA extracted from postmortem temporocortical tissue specimens (BA 41/42) belonging to 11 patient-control pairs, using the GeneChip Mapping Arrays 250K arrays (Affymetrix). Genome-wide expression was assessed in 6 of these 11 pairs, using the Human Genome 133 plus arrays (Affymetrix). SNP intensity data (log2 ratios) were analyzed using the CNAG software.

Results: false negative rates average 23% and 20%, while false positives are at 2.52% and 2.47% for nsp-array and sty-arrays, respectively. There is no significant difference between patients and controls in percentage of duplicated or deleted SNPs, although there appear to be genomic regions that are selectively duplicated/deleted in autistic patients or in controls. Expression profiling of the superior temporal gyrus of six autistic subjects and matched controls reveals increased transcript levels of many immune system related genes. CNV-gene expression correlations are being assessed.

Conclusions:The reliability of our CNV analysis appears quite satisfactory. No widespread increase in genomic instability is found in our postmortem brain sample of autistic patients. The genomic location of CNVs may be more relevant to autism vulnerability than their quantitative representation. Assessments of CNV-gene expression correlations will be important to understand the pathophysiology of CNV contributions to autism pathogenesis.