Researchers in the department of biostatistics and bioinformatics at Emory University Rollins School of Public Health have been awarded a four-year $1.5 million R01 grant from National Institute of Mental Health to develop novel statistical methods for integrating multi-dimensional data to explore mechanisms underlying mental disorders. Dr. Ying Guo, associate professor and acting director of the Center for Biomedical Imaging Statistics (CBIS), and Dr. Jian Kang, assistant professor and a core faculty member of CBIS, will lead a multidisciplinary team with experts from psychiatry, neuroscience and genetics to carry out the study.
Mental disorders are the leading cause of disability in the USA and roughly one out of 17 adults has a seriously debilitating mental illness. Many mental health studies now collect data from multiple platforms including neuroimaging, genetics, and behavioral sciences to investigate the biological mechanisms underlying mental disorders. This wealth of diverse datasets provides an unprecedented opportunity for crosscutting investigations that may offer new insights. For example, the integration of imaging and genetic data can potentially help reveal how genetic variants impact the neuronal function, leading to system-level dysfunction which then alters information processing in brain linked to mental illness.
However, integrative analysis of these diverse and multi-dimensional biological data is a challenging task. In their project, Drs Guo and Kang seek to develop a novel Distributional
Independent Component Analysis (D-ICA) method under the Bayesian modeling framework for tackling this problem. The proposed method can jointly decompose data from multiple platforms and extract integrated multi-dimensional profiles to facilitate diagnosis and deepen mechanistic understanding of mental disorders. Specifically, they will combine information from multimodal neuroimaging including functional MRI (fMRI), structural MRI (sMRI) and diffusion tensor imaging (DTI) to obtain more conclusive evidence on neural circuitry underlying mental disorders. Furthermore, they will develop computationally efficient methods to discover important linkage between genetic variants and brain functional networks. The identified imaging-genetic features can then be associated with clinical and behavioral symptoms to reveal how genetic variants alter the risk for mental disorders via brain phenotypes.
With recent advances in neuroimaging and genomics, mental health research has entered a new era with exciting opportunities to revolutionize diagnosis and treatment for mental disorders. The new statistical methods that Drs Guo and Kang propose will provide highly effective analytical tools to help achieve these goals.