Connectomic signatures advance the understandings of transdiagnostic heterogeneity in psychosis

Abstract

Current consensus-driven diagnostic systems poorly capture the marked heterogeneity and high comorbidity of psychiatric disorders, necessitating brain-based transdiagnostic frameworks to advance mechanistic understanding and clinical translation. Connectome-wide association studies have consistently identified widespread large-scale dysconnectivity across diagnostic boundaries, yet normative modeling and meta-analytic evidence highlights substantial inter-individual variability in the spatial topography and magnitude of connectomic deviations. In this Perspective, we bridge this divide between cross-disorder commonalities and individualized pathology by integrating connectomic overlap and computational biotyping into a top-down, four-tiered hierarchical architecture. This framework progresses from macro-scale transdiagnostic topological scaffolds, through meso-scale phenotypic triangulation, to data-driven biotypes and micro-scale clinical translations, delineating bidirectional interplay between shared network constraints and heterogeneous pathological diversification. Building on this architecture, we outline strategies leveraging multi-cohort data harmonization and healthy-anchored normative biotyping to refine robust cross-disorder circuits and reproducible neurobiotypes. Lastly, we propose a translational roadmap to transition connectomic stratification toward prospective, treatment-relevant clinical validation in precision psychiatry.

Abstract of "Connectomic signatures advance the understandings of transdiagnostic heterogeneity in psychosis" by Du Y, Zhang J, Ma T et al., https://doi.org/10.1038/s43856-026-01967-y, used under its Creative Commons Attribution license (CC-BY); markup removed, otherwise unmodified.