Abstract
Autism Spectrum Disorder (ASD) affects 1% of the population and it is characterized by social impairment and repetitive behaviors. Usually, signs of ASD occur before the age of three years and are often comorbid with other psychiatric disorders such as epilepsy and Intellectual Disability (ID) (25–70% of ASD cases have ID). ASD is believed to have a high degree of genetic heterogeneity with many rare variants contributing to the disorder. Thus far, most ASD research has focused on de novo and dominant variants because emphasis is placed on smaller Western families where recessive variants are hard to identify. This has led to a substantial knowledge gap on the effects recessive variants have on diseases such as ASD. The aim of this study is to assess the role recessive variants play in ASD and neurodevelopmental disorders.
We used Whole Exome Sequencing (WES) for a subset of samples so far. In addition, we have DNA from 257 Iranian, 115 Pakistani, and 20 Saudi ASD trios that will be analyzed over the next few months. The analysis pipeline consists of genome alignment using SNAP and the Genome Analysis Toolkit GlobalHaplotyper (GATK) implemented on a high performance Hadoop infrastructure. Variant calls are annotated using recently released databases and pathogenicity scores including M-CAP, Intervar, Sift, Polyphen2, MutationTaster and comparing against 123 000 exomes in gnoMAD and 2497 exomes in the GME Server.
We have identified de novo variants in known ID or ASD genes ZNF292, MYT1L, CHD8, DYRK1A, and novel genes such as ANXA4; also an X-linked variant in known ID gene THOC2. Biallelic mutations have been found in the known ID/ASD gene CC2D1A, and new candidate gene FAXDC2. The FAXDC2 is a homozygous truncating mutation that has not been associated with ASD or ID previously. FAXDC2 is a fatty acid hydroxylase domain containing protein with no known function or previous associations with disease. ANXA4 is part of the annexin family of genes previously associated with ID/ASD. It is a calcium dependent phospholipid binding protein and is expressed in the brain. Mouse models of ANX4 indicate behavioral and neurological impairments when knocked out.
We are in the process of analyzing 300 trios the results of which will be available in next few months. Once the function of these mutations and the associated clinical features is characterized, these genes can be utilized for molecular diagnosis, better clinical management and in the long-term may act as potential targets for personalized therapeutics.