Thesis & Dissertations

Copy Number Variants and Homozygosity Mapping in Autism in Lebanon


Introduction: Autism is one of the most mysterious and complex neurodevelopmental disorders. It is characterized by impairment in social and verbal interaction in addition to repetitive stereotyped behavior and restricted interests. In support of a strong underlying genetic predisposition is the higher prevalence in monozygotic twins versus fraternal twins. According to genetic studies on autism worldwide, a large number of genes are attributed to autism including genes within copy number variation (CNV) regions. In addition, CNVs in autism spectrum disorder (ASD) are much more numerous than in the non-autistic population. In our study we assessed CNVs and also compared homozygous regions specific to children with autism as compared to those of parents and siblings. Genes in these loss of heterozygosity (LOH) regions were compared to previously described autism, schizophrenia and mental disorder susceptibility genes (eg. CNTNAP2, GLO1, NRXN1 and GRIK2) in order to uncover genes in Lebanese children with autism previously described or to uncover novel autism susceptibility genes. Methods: DNA samples were extracted from peripheral blood of 37 children with autism as well as their family members using QIAamp┬« blood midi kit. CNV Screening was done using Affymetrix Cytogenetics 2.7M arrays as well as the CytoScan™ HD Arrays. Genes found in the CNV regions were then classified according to their biological processes in the human protein ontology database in the DAVID bioinformatics software (Database for Annotation, Visualization and Integrated Discovery). In addition, Pathway Studio software version 9.0 (Ariadne Genomics, Rockville, Md., USA) was used to construct pathways and identify altered protein interaction maps. The number of CNVs was counted for each family member and classified into inherited or de novo (autistic individuals and siblings). These were further classified into five groups according to database of genomic variants (DGV) database as being pathogenic, likely pathogenic, unknown, likely benign or benign. Results: The number of CNVs in children with autism were compared to the rest of family members as well as to non-autistic Lebanese controls. The number of CNVs in children with autism was more than that in non-autistic controls. In addition, the inherited CNVs in children with autism were greater in number than de novo CNVs. Also, the percentage of gains is greater with larger CNVs while that of losses is greater with smaller CNVs. These findings are in concordance with results of genetic studies done worldwide. Furthermore, the classification of genes in the CNV regions by DGV software revealed 2% pathogenic and 2% likely pathogenic inherited CNVs, in addition to 5% pathogenic and 5% likely pathogenic de novo CNVs. It is worthwhile to mention that in these CNV regions, genes already described in autism, such as CNTNAP2, were found in addition to novel candidate genes in the Lebanese that were related to the central nervous system. Conclusion: Uncovering a small set of genes responsible for autism in the Lebanese will facilitate diagnostic work-up and genetic counseling for autism spectrum disorders in Lebanon. Also, discovery of novel autism susceptibility genes will lead to potentially untapped therapeutic targets.


Sylva Garabet Kourtian


Mohamed Moustafa, Rose-Mary Na'Aman Boustany