Toward a fine-grained characterization of the intrinsic functional connectome in ASD

Background:   While the dysconnection model of autism has been increasingly supported by direct and indirect evidence from diverse fields of neuroscience, a fine-grained understanding of the underlying mechanisms has yet to emerge. This is a fundamental requirement to accelerate the translational identification of biomarkers and development of biologically-informed treatments. 

Objectives:   We surveyed a range of resting state fMRI measures capturing distinct signal properties for a comprehensive view of the functional connectome in ASD. 

Methods:   We examined 763 resting state functional MRI (R-fMRI) datasets (n=360 from individuals with ASD and n=403 from typical controls) available in the ABIDE repository, using several different analysis approaches. Specifically, 1) to examine local intrinsic functional connectivity (iFC) we computed Regional Homogeneity (ReHo); 2) to assess local information processing we employed Degree Centrality (DC) a graph-based measure of network organization; 3) to capture interhemispheric connectivity we calculated Voxel-Mirrored Homotopic Connectivity (VMHC). Finally, beyond connectivity, we examined the fractional amplitude of slow frequency fluctuations(fALFF) in the BOLD signal that underlies iFC. 

Results:   We examined 763 resting state functional MRI (R-fMRI) datasets (n=360 from individuals with ASD and n=403 from typical controls) available in the ABIDE repository, using several different analysis approaches. Specifically, 1) to examine local intrinsic functional connectivity (iFC) we computed Regional Homogeneity (ReHo); 2) to assess local information processing we employed Degree Centrality (DC) a graph-based measure of network organization; 3) to capture interhemispheric connectivity we calculated Voxel-Mirrored Homotopic Connectivity (VMHC). Finally, beyond connectivity, we examined the fractional amplitude of slow frequency fluctuations(fALFF) in the BOLD signal that underlies iFC. 

Conclusions:   This initial work conducted in a large sample provides a framework for investigations that take account of the regional as well as functional variation of connectivity properties in autism and can inform the dysconnection model of autism. The findings show that the use of diverse analytic approaches in R-fMRI can provide complementary evidence and a comprehensive picture of atypical functional connectivity in ASD, with some regional convergence that may direct the search for connectivity-based biomarkers.