Diagnosing autism spectrum disorder (ASD) is a challenging problem and is based purely on behavioral descriptions of symptomology (Diagnostic and Statistical Manual-5th Edition/ICD-10). Numerous limitations (e.g., informant discrepancies, lack of adherence to assessment guidelines) to current diagnostic practices have the potential to result in misdiagnosis of the disorder. Prior research provides evidence that structural and functional magnetic resonance imaging (MRI) data collected from individuals with ASD exhibit distinguishing characteristics that differ in neural patterns of the brain. Our proposed deep learning model ASD-DiagNet exhibits consistently high accuracy for classification of ASD brain scans from neurotypical scans. We have integrated traditional machine learning and deep learning techniques that allow us to isolate ASD biomarkers from MRI data sets. Our method, called Auto-ASD-Network, uses a combination of deep learning and support vector machines to classify ASD scans from neurotypical ones. Such interpretable models would help one to explain the decisions made by deep learning techniques leading to knowledge discovery for neuroscientists.