While modern diagnostic classification systems aim to nosologically structure psychiatric disorders, they poorly align with the genetic, neurobiological, and environmental heterogeneity observed in these disorders. This limitation has complicated the search for clinically useful biomarkers for diagnosis and treatment. Recent work on genetic and environmental contributions to mental health indicates that this heterogeneity stems from differential involvement of diverse biological pathways within and across diagnostic clusters. This complex interplay presents a many-to-many mapping problem in psychiatry, where distinct pathophysiological processes can lead to similar clinical symptoms. Here, we argue that disentangling these biological mechanisms requires development of process-specific biomarkers that could replace non-specific neuroimaging markers widely used in neuropsychiatric research. We further propose a framework for biomarker research that adopts a biologically informed perspective integrating the interactions between genes and the environment to address this problem. Such a multidimensional framework holds promise for developing biology-driven models of psychiatric disorders, enabling treatment strategies tailored to individual pathophysiology.
Dukart et al. (Thu,) studied this question.