Novel clustering of items from the Autism Diagnostic Interview‐Revised to define phenotypes within autism spectrum disorders

Abstract

Heterogeneity in phenotypic presentation of ASD has been cited as one explanation for the difficulty in pinpointing specific genes involved in autism. Recent studies have attempted to reduce the "noise" in genetic and other biological data by reducing the phenotypic heterogeneity of the sample population. The current study employs multiple clustering algorithms on 123 item scores from the Autism Diagnostic Interview-Revised (ADI-R) diagnostic instrument of nearly 2000 autistic individuals to identify subgroups of autistic probands with clinically relevant behavioral phenotypes in order to isolate more homogeneous groups of subjects for gene expression analyses. Our combined cluster analyses suggest optimal division of the autistic probands into 4 phenotypic clusters based on similarity of symptom severity across the 123 selected item scores. One cluster is characterized by severe language deficits, while another exhibits milder symptoms across the domains. A third group possesses a higher frequency of savant skills while the fourth group exhibited intermediate severity across all domains. Grouping autistic individuals by multivariate cluster analysis of ADI-R scores reveals meaningful phenotypes of subgroups within the autistic spectrum which we show, in a related (accompanying) study, to be associated with distinct gene expression profiles.

Exploring the Relationship Between Autism Spectrum Disorder and Epilepsy Using Latent Class Cluster Analysis

“…Given that ASD and epilepsy could potentially have common genetic determinants, it is surprising that more studies have not examined the patterns of co-occurrence of other behaviors to identify potential ASD-epilepsy phenotypes (Tuchman et al 2009). To date there have been a number of studies attempting to identify distinct clusters within ASD datasets (Eaves et al 1994;Sevin et al 1995;Stevens et al 2000;Hu and Steinberg 2009). However, these cluster based studies have not investigated epilepsy rates within the various clusters.…”

A machine learning approach for identifying subtypes of autism

“…Recently, the National Institute of Health (NIH) relied on a machine learning program in an attempt to categorize ASD [4]. The researchers used the Autism Diagnostic InterviewRevised (ADI-R), and their goal was to correlate genetic markers with specific subgroups.…”

Subphenotype-Dependent Disease Markers for Diagnosis and Personalized Treatment of Autism Spectrum Disorders

“…In order to reduce the heterogeneity within the ASD population for genome-wide gene expression analyses, we developed a novel phenotyping method involving multivariate cluster analyses of severity scores from the ADI-R diagnostic instrument [15] (widely considered to be a "gold standard" behavioral diagnostic measure for ASD) to separate individuals with ASD into subgroups according to similarity of symptomatic behavioral profiles [16]. In contrast to the cited studies that divided the ASD population according to a single or a few specific traits (such as regression, repetitive behaviors, or loss of spoken words), we selected 123 item scores that captured a broad spectrum of behaviors manifested by individuals with ASD in order to identify phenotypic subgroups of individuals with idiopathic ASD on the basis of similarity of symptom severity across multiple functional domains.…”

“…GWA data was derived from the study of Wang et al [30] and ADI-R scoresheets for a subset of the genotyped individuals were obtained through the Autism Genetic Resource Exchange (courtesy of Dr. Vlad Kustanovich). The item scores from the ADI-R diagnostic measures were used in two ways: 1) to perform quantitative trait association analyses for 5 distinct autistic traits; and 2) to subtype individuals with ASD according to our published cluster analyses of ADI-R scores [16]. For quantitative trait association analyses, 14-23 ADI-R scores were summed to obtain each individual's cumulative score for each of the following 5 traits: language impairment, nonverbal communication, play skills, insistence on sameness and ritualistic behavior, and social development.…”