Metastability as a candidate neuromechanistic biomarker of schizophrenia pathology

• DOI: 10.1371/journal.pone.0282707
• Publication Date: 2023-03-23T18:09:46Z
• Authors: Fran Hancock; Fernando E. Rosas; Robert A. McCutcheon; Joana Cabral; Ottavia Dipasquale; Federico E. Turkh...
• Keywords: Biomarker Discovery

The disconnection hypothesis of schizophrenia proposes that symptoms the disorder arise as a result aberrant functional integration between segregated areas brain. concept metastability characterizes coexistence competing tendencies for and segregation in brain, is therefore well suited study schizophrenia. In this study, we investigate candidate neuromechanistic biomarker pathology, including demonstration reliability face validity. Group-level discrimination, individual-level classification, pathophysiological relevance, explanatory power were assessed using two independent case-control studies schizophrenia, Human Connectome Project Early Psychosis (HCPEP) (controls n = 53, non-affective psychosis 82) Cobre 71, cases 59). work extend Leading Eigenvector Dynamic Analysis (LEiDA) to capture specific features dynamic connectivity then implement novel approach estimate metastability. We used non-parametric testing evaluate group-level differences naïve Bayes classifier discriminate from controls. Our results show our new capable discriminating controls with elevated effect sizes relative published literature, reflected an up 76% area under curve (AUC) out-of-sample classification analyses. Additionally, metric showed 81–92% measures segregation. Furthermore, analyses demonstrated patients early exhibit intermittent disconnectivity subcortical regions frontal cortex cerebellar regions, introducing insights about mechanistic bases these conditions. Overall, findings demonstrate validity pathology.