Autism is currently defined as a single disorder characterized by impairments in social communication and the presence of restricted interests and repetitive behaviors. However, there is mounting evidence that autism represents a collection of overlapping neurodevelopmental disorders, resulting from a variety of neuroanatomical, neurophysiological, neuroimmunologic and/or genetic abnormalities. The wide range of possible etiologies and the heterogeneity of symptom expression among individuals with autism leads to speculation that there actually are several "autisms", i.e., clinically distinct disorders with similar behavioral presentations but different etiologies, clinical course, and treatments. The PDN research program aims to characterize the behavioral and biological manifestations of these differing presentations of autism and to identify their unique features in order to facilitate development of effective therapeutic and preventive interventions. During the reporting period, this project focused primarily on continuing a longitudinal, phenotyping investigation of 105 young children (ages 1-6 years) with autism, 60 age and sex-matched typically developing controls and a group of 25 children with non-autistic developmental delays (Protocol 06-M-0102, NCT 00298246). Baseline evaluations have been completed for these subjects and they are returning for periodic follow-up assessments. Data from the baseline evaluations are now being analyzed, including examinations of potentially meaningful subgroups of patients, such as the comparisons of children with autism who have a history of developmental regression and those without. Interestingly, one of the most striking findings of the preliminary analyses was that regression appears to be a continuous variable, rather than the dichotomous categories previously described. That is, children with a history of developmental regression have frequently had at least some delays in their early development and those with distinct early developmental delays may also lose some social and communication skills. We have also recently explored the measurement of restrictive and repetitive behavior in this cohort, finding that these children are reported to have significantly more restrictive and repetitive behavior than found in similar samples, and that this behavior remains stable over a one year period. Finally, data from this project have also been used to continue validation of appropriate algorithms of diagnostic instruments for children in the preschool age range. In addition to the behavioral assessments, the study includes comprehensive medical and developmental histories; neuropsychological, medical and neurological evaluations; assays of blood, urine and cerebrospinal fluid (CSF) samples; and also a variety of specialized studies, including routine and overnight electroencephalograms(EEGs); modified polysomnography to evaluate sleep architecture; and magnetic resonance imaging (MRI scans); genetic assays; and dysmorphology evaluations. Preliminary results from these evaluations have demonstrated that more than one-half of the children have abnormalities of sleep architecture, particularly notable are relative reductions in the percentage time spent in Rapid Eye Movement (REM) sleep. The finding prompted a therapeutic trial which is described in a separate project report (MH002914-04 PDNB). We also analyzed cross-sectional data from MRI scans, which found signals of abnormal myelination in specific brain areas of autism, as well as other indicators of abnormalities in brain white matter and relative cortical thinning in specific key regions in the brain. Evidence of previously reported abnormalities of early brain growth was also investigated, both through imaging as well as head circumference records, with data indicating previous findings may have spuriously overestimated increased head size in autism due to inadequate norms (see Raznahan et al, 2013). As expected, genetic abnormalities were found in approximately 10% of the subjects and the aberrations are being evaluated for clinical significance. Further, we are engaged in a collaboration to conduct whole exome sequencing of the cohort, and are in the process of analyzing the relationship of sequencing data and phenotypic variations. More information about this study (Protocol 06-M-0102, NCT 00298246) may be found at http://clinicalstudies.info.nih.gov/detail/A_2006-M-0102.html Other phenotyping studies are also underway or have been recently completed. Each of them shares the goal of providing a deeper understanding of the autism phenotype and its relationship to etiology, clinical course and outcome. The first of these is a prospective, longitudinal investigation of toddlers considered to be at-risk for ASD because of the presence of early language delays (Protocol 11-M-0144, NCT01339767). This study aims to delineate early communicative impairments that predict ASD and to distinguish these from non-specific markers of (non-autistic) developmental delays. The investigation also examines the relationship of communication impairments to abnormalities of brain structure and function assessed by electroencephalograms (EEGs), structural MRIs, and functional MRI scans. The longitudinal assessments also include comprehensive behavioral assessments designed to profile strengths and weaknesses in communication and other domains. The goal of these assessments is to identify specific risk and resilience factors for ASD. More information about this study may be found at: http://www.clinicaltrials.gov/ct2/show/NCT01339767?term=toddlers+autism&rank=2 A related investigation uses near-infrared spectroscopy (NIRS) to perform functional neuroimaging in children who are too young or behaviorally-challenged to tolerate functional MRI scanning procedures -- a key group that is missing from much of the neuroimaging literature on autism. Using tasks of inhibition and including comparison groups of children with attention-deficit hyperactivity disorder and typically developing children, this study seeks to obtain reliable and informative brain activation data on children with autism as young as 4 years old. A final area of phenotyping investigation involves behavioral assessments of individuals with specific genetic disorders purported to be highly associated with autism, such as Phelan McDermid Syndrome, Smith-Lemli-Opitz Syndrome and others. Additionally, through a collaboration with the Undiagnosed Diseases Program of NHGRI, subjects with serious rare and potentially genetic disorders are evaluated to find common and unique patterns of developmental and behavioral strengths and weaknesses. The behavioral assessments are conducted as part of PDN's "Natural History" investigation (Protocol 13-M-0028, NCT01778504), which is described at: http://clinicalstudies.info.nih.gov/cgi/detail.cgi?A_2013-M-0028.html