Patient Trust in Healthcare Providers among People on Methadone Maintenance Treatment Using Methamphetamine in Vietnam.

Identifying the genetic basis of the human brain’s complex network structure is essential for understanding how the brain develops, functions, and relates to health and cognition. Over a decade of work, rooted in seminal twin studies, has shown that individual differences in brain network organisation measured by multiple modalities of magnetic resonance imaging (MRI) are under genetic influence. Recent genome-wide association studies (GWAS) have since enabled the investigation of common genetic variation associated with certain brain network phenotypes, including (f)MRI-derived functional connectivity (FC) from correlational analysis of regional time-series, and diffusion (d)MRI-derived structural connectivity (SC) from tractographic reconstruction of axonal projections between regions. However, our understanding of genetic influences on brain network organisation remains fragmented. For example, it has proven difficult to identify any meaningful genetic overlap between brain structural and functional connectivity phenotypes derived from different MRI modalities. The spatially varying genetic architecture of localised (regional) MRI measures of brain structure has been well characterised by both twin studies and more contemporary GWAS. It has also been shown by both twin studies and GWAS that common genetic effects are largely distributed across major systems or classes of cortical areas, implying that formation of large-scale cortical gradients (of lamination) or cortical networks (of axonal or synaptic connectivity) are under genetic control. However, it has not yet been shown how such genetically-driven cortical gradients derived from structural MRI data can be related to prior theories of cortical organisation, or measures of functional connectivity, or genetic risks for neuropsychiatric disorders and biomedical traits. To address these key knowledge gaps, we used an innovative approach to human MRI network phenotyping, based on estimating the structural similarity between cortical areas. It has already been established that the similarity between cortical areas, which is measurable at the whole-brain scale of MRI, e.g., in terms of inter-areal correlation of macro- and/or micro-structural MRI feature vectors, is representative of the architectonic similarity between areas at a cellular scale. Architectonically similar areas of cortex are also more likely to be inter-connected by axonal projections. As such, structural similarity is well positioned to bridge structural MRI features defined locally (e.g., cortical thickness) with network-based measures that are defined relationally between brain areas (e.g., functional connectivity). Several methods have emerged recently to estimate the connectome from MRI similarity analysis. We used Morphometric INverse Divergence (MIND) for several technical, biological, and practical reasons. First, it has been biologically validated and technically benchmarked favourably against alternative structural connectivity and similarity estimators Importantly, MIND similarity was strongly co-located with cortical patterns of genomic co-expression, and has moderate twin-based heritability (average nodal ), indicating that MIND phenotypes might be advantageous for analysis of genetic effects on brain networks. Finally, since its recent development, MIND has been rapidly taken up by the research community for the analysis of brain network organisation across many different applications, including lifespan development, neuropsychiatric and neurodegenerative disorders, cross-species connectomics, and beyond. As such, a comprehensive genetic analysis of MIND network phenotypes is not only empirically and theoretically justified, but also timely.