Network level analysis for connectome-wide association studies: applications in the developing brain

Abstract: Cutting-edge brain connectivity mapping techniques, including diffusion and functional MRI, in combination with NIH-funded initiatives such as the Human Connectome Project and NIH Toolbox, have accelerated production of large, highly valuable datasets that combine rich neuroimaging and behavioral assessments. As these datasets are established, innovative analysis approaches are required to delineate reliable links between variability in brain connectivity (i.e., the connectome) and behavior (e.g., measures of cognition, developmental changes, and disease). Current analytic methods often employ highly stringent false positive rate (FPR) corrections that yield sparse, scattered brain-behavior associations, challenging biological interpretation and reproducibility. The Wheelock lab develops statistical software to analyze connectome-wide associations while controlling the FPR by leveraging the inherent systems level organization of brain networks. However, challenges remain for defining these systems in early childhood (i.e., less than 3 years of age). Ongoing research in the lab seeks to benchmark reproducibility and determine best approaches for determining connectome-wide associations with clinical outcomes in early development. 

Network level analysis for connectome-wide association studies: applications in the developing brain

Abstract: Cutting-edge brain connectivity mapping techniques, including diffusion and functional MRI, in combination with NIH-funded initiatives such as the Human Connectome Project and NIH Toolbox, have accelerated production of large, highly valuable datasets that combine rich neuroimaging and behavioral assessments. As these datasets are established, innovative analysis approaches are required to delineate reliable links between variability in brain connectivity (i.e., the connectome) and behavior (e.g., measures of cognition, developmental changes, and disease). Current analytic methods often employ highly stringent false positive rate (FPR) corrections that yield sparse, scattered brain-behavior associations, challenging biological interpretation and reproducibility. The Wheelock lab develops statistical software to analyze connectome-wide associations while controlling the FPR by leveraging the inherent systems level organization of brain networks. However, challenges remain for defining these systems in early childhood (i.e., less than 3 years of age). Ongoing research in the lab seeks to benchmark reproducibility and determine best approaches for determining connectome-wide associations with clinical outcomes in early development.