Brain Structural And Functional Signatures Of White Matter In Major Depressive Disorder

Depression is a multifactorial disorder with clinical heterogeneous features characterized by persistent depression,anxiety and agitation,changes in social behavior,and sleep abnormalities.The incidence of depression is increasing every year and is associated with a high risk of death.No less than 60% of people with depression have more than one relapse,which leads to a serious social and economic burden.Although the emergence of functional MRI provides supports for exploring neurological and psychiatric disorders,the pathophysiological mechanisms of depression remain unclear.Therefore,the search for neuroimaging-based biomarkers for diagnosis and treatment of depression is a current imperative.fMRI is a non-invasive technique that allows for visualization of activated brain regions and has been widely used in various scientific research and clinical practice,especially in psychiatric disorders.However,previous f MRI studies have focused on gray matter and ignored the functional information contained in the WM of the brain.A growing number of studies in recent years have confirmed the validity of functional information in the WM of the brain,which has greatly expanded the breadth of brain connectivity research,as well as improve the assessment and diagnosis of connectivity disorders.However,previous brain imaging studies have tended to focus on singlemodality imaging data depicting very limited information about the brain,and in addition to investigating the independent role of structure and function in depression,studies also need to consider the relationship between structure and function.Therefore,this thesis mainly used magnetic resonance imaging techniques,including blood-oxygenation-leveldependent f MRI and diffusion tensor imaging,combined with brain gene expression data,aiming to explore abnormal brain network representations in depression from different dimensions and provide a new perspective to explore the neuropathological mechanisms of depression.The main contents are as follows:Firstly,the thesis constructed a new method that combined structural connectivity with functional connectivity in WM portrays structure-function coupling by calculating Pearson correlation coefficients between structural and functional indicators to detect the abnormal distribution pattern of the WM network in depression.Then,structural,functional,and structural-functional coupling indicators were used as features to construct prediction models of Hamiltonian scores and classification models of healthy controls/depression,respectively.It was found that compared to healthy controls,depressed patients showed a pattern of globally reduced WM microstructural integrity,while WM networks with abnormal functional connectivity was somewhat consistent with WM networks with abnormal structural abnormalities.Moreover,WM networks with significant abnormalities in both structure and function also showed abnormal organization patterns in structure-function coupling.Structure-function coupling always showed better performance in both depression severity prediction models and in classification models of depressed patients and healthy subjects.These findings suggested that structure-function coupling has a higher sensitivity to disease-induced changes and could detect a more comprehensive manifestation of abnormalities.Then,using the imaging data from f MRI,a method to detect individual variability was proposed to explore the spatial distribution pattern of individual variability in depression.To investigate the relationship between individual variability in functional connectivity within the WM network of depression and cognition and behavior,this thesis performed a meta-analysis of the probability distribution patterns obtained from WM network mapping.It was observed that individual variability of functional connectivity in WM network localization outcomes in depression was associated with both primary sensorimotor as well as cognitive functions,which findings support previous studies that revealed abnormal sensorimotor abilities as well as impaired cognitive abilities in depression.Further,the mechanism of gene expression closely related to this spatial pattern of individual functional connectivity individual variability was further explored by combining micro-scale gene expression data.It was revealed that WM functional connectivity variability exhibited lower global means as well as higher standard deviations in depression.Using Allen brain genetic data,gene expression weights of PLS1 were associated with abnormal patterns of individual variability in depressed patients,and that these genes were mostly enriched in pathways that related to cells.These results suggested that combining information from different dimensions of brain networks may deepen the understanding of the underlying molecular mechanisms of depression.In conclusion,this thesis combined multimodal imaging data and Allen brain gene expression data to construct a structure-function model and a brain image-gene expression model,respectively,to reveal the structural and functional abnormalities of brain networks and the alterations in WM in this pathological condition of depression were analyzed from different dimensions,which supports the observation of the pathological mechanisms of depression.