Abstract:BackgroundIn the United States, about 3 million people have autism spectrum disorder (ASD), and around 1 out of 59 children are diagnosed with ASD. People with ASD have characteristic social communication deficits and repetitive behaviors. The causes of this disorder remain unknown; however, in up to 25% of cases, a genetic cause can be identified. Detecting ASD as early as possible is desirable because early detection of ASD enables timely interventions in children with ASD. Identification of ASD based on objective pathogenic mutation screening is the major first step toward early intervention and effective treatment of affected children. ObjectiveRecent investigation interrogated genomics data for detecting and treating autism disorders, in addition to the conventional clinical interview as a diagnostic test. Since deep neural networks perform better than shallow machine learning models on complex and high-dimensional data, in this study, we sought to apply deep learning to genetic data obtained across thousands of simplex families at risk for ASD to identify contributory mutations and to create an advanced diagnostic classifier for autism screening. MethodsAfter preprocessing the genomics data from the Simons Simplex Collection, we extracted top ranking common variants that may be protective or pathogenic for autism based on a chi-square test. A convolutional neural network–based diagnostic classifier was then designed using the identified significant common variants to predict autism. The performance was then compared with shallow machine learning–based classifiers and randomly selected common variants. ResultsThe selected contributory common variants were significantly enriched in chromosome X while chromosome Y was also discriminatory in determining the identification of autistic individuals from nonautistic individuals. The ARSD, MAGEB16, and MXRA5 genes had the largest effect in the contributory variants. Thus, screening algorithms were adapted to include these common variants. The deep learning model yielded an area under the receiver operating characteristic curve of 0.955 and an accuracy of 88% for identifying autistic individuals from nonautistic individuals. Our classifier demonstrated a considerable improvement of ~13% in terms of classification accuracy compared to standard autism screening tools. ConclusionsCommon variants are informative for autism identification. Our findings also suggest that the deep learning process is a reliable method for distinguishing the diseased group from the control group based on the common variants of autism.

Common genetic risk variants identified in the SPARK cohort support DDHD2 as a candidate risk gene for autism

Diagnostic Classification and Prognostic Prediction Using Common Genetic Variants in Autism Spectrum Disorder: Genotype-Based Deep Learning

Detection of Clinically Relevant Genetic Variants in Autism Spectrum Disorder by Whole-Genome Sequencing

Common risk variants identified in autism spectrum disorder

Common genetic variants on 5p14.1 associate with autism spectrum disorders

Inherited and De Novo Genetic Risk for Autism Impacts Shared Networks

Genomic analysis of 116 autism families strengthens known risk genes and highlights promising candidates

Genome-wide Analyses of Exonic Copy Number Variants in a Family-Based Study Point to Novel Autism Susceptibility Genes.

Genome‐wide Association Analysis of Autism Identified Multiple Loci That Have Been Reported As Strong Signals for Neuropsychiatric Disorders

eQTL colocalization analysis highlights novel susceptibility genes in Autism Spectrum Disorders (ASD)

INHERITED AND DE NOVO GENETIC RISK FOR AUTISM IMPACTS SHARED BIOLOGICAL NETWORKS

Noncoding de novo mutations in are associated with autism spectrum disorders

Common variants on 1p13.2 associate with risk of autism.

Discovery and validation of novel genes in a large Chinese ASD cohort

Analysis of recent shared ancestry in a familial cohort identifies coding and noncoding autism spectrum disorder variants

[Analysis of Common Genetic Variants Associated with Neuro-Synapse Development among 60 Family Trios Affected with Sporadic Autism Spectrum Disorders].

Common Genetic Variants on 1p13.2 Associate with Risk of Autism

DIP2C polymorphisms are implicated in susceptibility and clinical phenotypes of autism spectrum disorder

Shared Genetic Determinants of Schizophrenia and Autism Spectrum Disorder Implicate Opposite Risk Patterns: A Genome-Wide Analysis of Common Variants

Convergent Coexpression of Autism-Associated Genes Suggests Some Novel Risk Genes May Not Be Detectable in Large-Scale Genetic Studies.

Discovery and Validation of Novel Genes in a Large Chinese Autism Spectrum Disorder Cohort.