Developmental brain disorders (DBD) comprise an etiologically-heterogeneous, behaviorally-defined group of conditions affecting a significant percentage of U.S. children and adults. Shared genomic underpinnings now directly connect seemingly unrelated DBD, including autism (ASD), intellectual disability and schizophrenia. DBD have traditionally been studied using categorical criteria (e.g., ASD versus no ASD) that are no longer consistent with biological evidence. Here, we propose a genome-first approach to the study of DBD, examining normally distributed quantitative traits that extend into the general population. We will characterize differential patterns of functional and medical impact of DBD-related pathogenic loss of function (pLOF) variants in 100,000 patients from Geisinger Health System through the use of EHR data mining algorithms, with the following aims: 1) Use a cross-disorder, genomics-driven approach to create a comprehensive knowledge base of DBD genes and copy number variant (CNV) regions. We will expand the knowledge base for identifying DBD loci using existing tools from our previous work: 1) a dosage map of the human genome and 2) a loss of function, single gene prediction model based on literature annotation for six specific types of DBD. High-confidence DBD genes and CNV regions will be used in Aim 2 for genomic variant mining of whole exome sequencing (WES) data. 2) Conduct large-scale, population-based analyses to identify a cohort with DBD pLOF variants. We will use WES data, generated as part of another Geisinger study, to analyze the frequency of DBD pLOF variants in 100,000 patients representing an unselected cross-section of our patient population. Individuals with a DBD pLOF variant (n=4,000) and their family members will be phenotyped in Aims 3 & 4. 3) Characterize quantitative phenotypes in individuals with DBD pLOF variants and their families. Among patients with DBD pLOF, we will use linked EHR data to document clinical DBD diagnoses. We will administer in-person and online assessments to measure quantitative, heritable traits including cognitive performance, adaptive behavior, and social function. We will also assess these traits in first-degree relatives to examine the effect of family background on phenotypic variability. We will describe dimensional, quantitative neurodevelopmental profiles that capture the full phenotypic spectrum associated with specific pLOF variants. 4) Assess medical comorbidities and healthcare utilization among individuals with DBD pLOF variants. We will survey our EHR dataset to compare medical comorbidities and healthcare utilization between individuals with and without pLOF variants. We will examine a range of variables including DBD-related specialty visits and prescription use, hospitalizations and surgical procedures, hypothesizing that comorbidities and utilization will be significantly higher among people with pLOF variants, regardless of whether they have a known clinical DBD diagnosis. Characterization of differential patterns of clinical diagnosis, neurodevelopmental function, medical impact and healthcare utilization based on genomic variants will ultimately pave the way for targeted interventions for DBD.
Interagency Autism Coordinating Committee (IACC)
Autism Research Database (AFD)
