Towards Molecular Diagnosis of Autism Spectrum Disorder Using Multi-Omics Approaches : The Baraka-Qatar Study

Autism spectrum disorder (ASD) is a neurodevelopmental condition characterized by impaired social and communication skills, restricted interests, and repetitive behaviors. The prevalence of ASD among children in Qatar was recently estimated to be 1.1%, though the genetic architecture underlying ASD both in Qatar and the greater Middle East has been largely unexplored. In this study, we describe the first genomic and metabolomic data release from the BARAKA-Qatar Study – a nationwide program building a broadly consented biorepository of children with ASD and their families available for multi-omics research. For genomic data (Chapter 3), we present a comprehensive analysis of whole-genome sequencing (WGS) data of the first 100 families (372 individuals), investigating the genetic architecture, including single-nucleotide variants (SNVs), copy number variants (CNVs), tandem repeat expansions (TREs), as well as mitochondrial variants segregating with ASD in local families. In total, we discover potentially pathogenic variants in known genes/regions in 27 of 100 families (27%). Dominant risk variants, including de novo and inherited variants, contributed to 15 (55.6%) of these families, where the majority of dominant risk factors derived from SNVs/indels (66.7%) compared to CNVs (13.3%), TREs (13.3%), and mtDNA variants (6.7%). Moreover, recessive risk of homozygous variants was found in 7 (25.9%) families with an almost 6-fold increase in homozygous causative variants in consanguineous versus non-consanguineous families (13.6% and 1.8% respectively). In addition, X-linked damaging missense variants were identified in five male probands (18.5%). According to ACMG guidelines, 11 of the 27 discovered variants (40.7%) were classified as pathogenic and likely-pathogenic. We also searched genome-wide for novel ASD candidate genes/regions and identified 32 novel genes with genetic variants in 32 of 100 families (32%). Of these novel genes, 27 out of 32 (84.4%) genes were supported by additional carriers affected by variants in similar classes and zygosity in other ASD cohorts including MSSNG, SSC, and SPARK.

In addition to genomic analysis, we generated metabolomic data on 58 cases and their control siblings to investigate pathways perturbed in ASD (Chapter 4). We employed untargeted metabolomics using ultrahigh-pressure LC-MS to maximize the coverage of plasma metabolome. We discovered 253 metabolites dysregulated in cases with ASD, enriched in four biological pathways, these include the urea cycle, glutamate metabolism, ammonia recycling, and arginine and proline metabolism. The impacted pathways observed in our results overlapped with the results of previous studies. Our study represents a systematic approach for emerging ASD studies in under-represented populations and provides a unique resource of Middle Eastern genomes for future research to the global ASD community and highlights the importance of WGS as a comprehensive tool in providing a molecular diagnosis for families with ASD. Moreover, we uncover a significant role for recessive variation in ASD architecture in consanguineous settings. Lastly, we explore how integrating other omics data such as metabolomics can reveal complex biological pathway interactions occurring in individuals with ASD.