Combining systems biology and big data analytics to uncover metabolic and environmental factors in autism spectrum disorder

Abstract

Given the inherent complexity of ASD, reaching this goal will necessitate a deviation from traditional medical approaches, which typically focus on single physiological mechanisms individually, towards systems-oriented and big data methodologies to investigate these systems concurrently. This work will begin with the development of a mathematical model for capturing folate-dependent one-carbon metabolism and transsulfuration activity in individuals with ASD. Estimation of personalized model parameters using clinical case-control data and calculation of parameter distributions will reveal metabolic reactions where abnormalities may be present in ASD. The next area of investigation will use multivariate statistical methods to evaluate the efficacy of biochemical measurements as potential markers for diagnosing ASD and evaluating outcomes of clinical treatment. Results will be assessed in the context of the models’ abilities to predict new data that were not used during model development. Finally, a retrospective analysis of administrative medical claims data will be performed to examine patterns in the diagnosis of co-occurring conditions in children with ASD. Studying co-occurring conditions and their contributing factors offers potential for understanding the underlying mechanisms of subgroups of ASD, which can be explored in future work.

Autism spectrum disorder (ASD) is estimated to affect 1 in 59 children in the United States. With the etiology of ASD still under general debate, the current standards for diagnosis are behavioral evaluations based on clinical observation or parent reporting. However, these evaluations are inherently subjective and do not offer an unbiased assessment of ASD status that a biomarker can offer. A common consequence of this gap in knowledge is an ASD diagnosis that is delayed by several years, which also delays behavioral interventions that could contribute to improved outcomes for individuals with the disorder. Achieving greater diagnostic accuracy along with improved outcomes for individuals with ASD will therefore require a more complete understanding of the metabolic and environmental factors contributing to the pathophysiology of the disorder.