Enhancing studies of the connectome in autism using the autism brain imaging data exchange II

Abstract

The second iteration of the Autism Brain Imaging Data Exchange (ABIDE II) aims to enhance the scope of brain connectomics research in Autism Spectrum Disorder (ASD). Consistent with the initial ABIDE effort (ABIDE I), that released 1112 datasets in 2012, this new multisite open-data resource is an aggregate of resting state functional magnetic resonance imaging (MRI) and corresponding structural MRI and phenotypic datasets. ABIDE II includes datasets from an additional 487 individuals with ASD and 557 controls previously collected across 16 international institutions. The combination of ABIDE I and ABIDE II provides investigators with 2156 unique cross-sectional datasets allowing selection of samples for discovery and/or replication. This sample size can also facilitate the identification of neurobiological subgroups, as well as preliminary examinations of sex differences in ASD. Additionally, ABIDE II includes a range of psychiatric variables to inform our understanding of the neural correlates of co-occurring psychopathology; 284 diffusion imaging datasets are also included. It is anticipated that these enhancements will contribute to unraveling key sources of ASD heterogeneity.

Figure 1. Key phenotypic characteristics.

(a) Total number of datasets per group (gray=controls; blue=autism spectrum disorder (ASD)) for the 17 cross-sectional ABIDE II data collections (i.e., collections from individuals not included in ABIDE I). Data are ordered as a function of sample size. (b) Number of males (light blue) and females (red) for each data collection, irrespective of diagnostic group. Data are ordered as a function of sample size. (c) Age at time of scan in years per collection (ordered by mean age per collection), irrespective of diagnostic group. The median age across collections (11.7 years) is depicted with a thick red dashed line; 25th, 75th, and 90th percentiles (9.3, 18.6, and 25.5 years, respectively) are represented by thin red dashed lines. (d) Distribution of full scale IQ (FIQ) standard scores per collection (ordered by lowest FIQ included per collection) for all datasets, irrespective of diagnostic group. The median FIQ across collections (112) is depicted with a thick red dashed line; 25th, 75th, and 90th percentiles (101, 122, and 130, respectively) are represented by thin red dashed lines. (e) Tukey’s box-whiskers plots depict the distribution of Autism Diagnostic Observation Schedule, Second Edition (ADOS-2) total calibrated severity scores (CSS) for ASD datasets in the nine collections sharing them (ordered by mean CSS per collection). The black plus sign depicts the mean CSS for each collection. (f) Distribution of Social Responsiveness Scale (SRS) total T scores (gray=controls; blue=ASD) in the 12 collections sharing them. For each collection, red dashed and solid lines indicate mean SRS total T scores of ASD and controls, respectively.

Figure 2. Data on psychiatric comorbidity in Autism Spectrum Disorder (ASD).

(a) Number of ASD datasets with and without psychiatric diagnoses for each of the nine data collections sharing information on categorical psychiatric diagnoses other than ASD. (b) Percentage of ASD datasets with one (black), two (dark gray) or more (light gray) comorbid diagnoses and those without any comorbidity (white) across the nine collections sharing comorbidity information. (c) Distribution of ASD datasets with comorbidity divided into those with comorbid Attention-Deficit/Hyperactivity Disorder (ADHD, green), anxiety (blue/white pattern) or others (cyan-white pattern) for each collection. Other comorbidities include enuresis, and/or mood, speech and language and/or disruptive behavior disorders. Here, given that multiple comorbid disorders can co-occur, the number of comorbid ASD datasets across categories exceeds the absolute number of comorbid ASD datasets.

Figure 3. Selection of spatial quality assurance (QA) metrics for high resolution MRI datasets.

(a) Contrast-to-noise ratio (CNR), (b) smoothness of voxels indexed as full half-width maximum (FHWM), (c) signal-to-noise ratio (SNR), (d) artifactual voxel detection (Q_i1)- See Table 5 for details on this and the other quality metrics released. The colored scatterplots illustrate the quality metrics distribution for spatial MRI dataset within a given ADBIE II collection (17 cross-sectional and 2 longitudinal collections). The black and white violin plots represent a kernel density estimation of the distribution across all datasets for each quality metrics. The midline thick gray line represents the value that occurs most commonly in the distribution. For each plot the horizontal gray lines mark the 1st, 5th, 25th, 50th (solid gray line), 75th, 95th and 99th percentiles starting from the bottom.

Figure 4. Selection of spatial and temporal quality metrics for resting state functional MRI (R-fMRI).

Spatial metrics include: (a) Ghost to single ratio (GSR); (b) smoothness of voxels indexed as full-width half maximum (FWHM), (c) signal to noise ratio (SNR). Temporal metrics are: (d) mean framewise displacement; (e) standardized DVARS, and (f) global correlation (GCORR)—See Table 5 for details on this and the other quality metrics released. The colored scatterplots illustrate the quality metrics distribution for spatial MRI dataset within a given ADBIE II collection (17 cross-sectional and 2 longitudinal collections). The black and white violin plots represent a kernel density estimation of the distribution across all datasets for each quality metrics with its midline thick gray line representing the value that occurs most commonly in the distribution. For each plot, the horizontal gray lines mark the 1st, 5th, 25th, 50th (solid gray line), 75th, 95th and 99th percentiles starting from the bottom.