# Integrative ceRNA network analysis in monozygotic twins reveals shared and disorder-specific molecular signatures in major psychiatric disorders

## TL;DR

This study identifies shared and disorder-specific RNA regulatory networks in psychiatric disorders using twin data, offering insights into molecular mechanisms and potential diagnostic targets.

## Contribution

The study reveals disorder-specific and shared ceRNA networks in psychiatric disorders using twin data and integrative transcriptomic analysis.

## Key findings

- 19 miRNAs are shared across schizophrenia, bipolar disorder, and major depressive disorder, with disorder-specific lncRNAs and mRNAs.
- Extracellular matrix-receptor interaction pathways are convergently involved in psychiatric disorders through ceRNA networks.
- COL6A6 and ITGB8 show strong predictive potential in blood and brain regions for specific disorders.

## Abstract

Psychiatric disorders such as schizophrenia (SCZ), bipolar disorder (BD), and major depressive disorder (MDD) share overlapping features but arise from distinct molecular mechanisms. Competitive endogenous RNA (ceRNA) networks, where long non-coding RNAs (lncRNAs) and mRNAs compete for shared microRNAs (miRNAs), represent a key regulatory layer. This study sought to identify disorder-specific and convergent ceRNA regulatory signatures across these conditions.

We constructed an integrative analysis of whole-transcriptome and small RNA sequencing data from peripheral blood samples of monozygotic twin pairs discordant for disease. Differentially expressed mRNAs, lncRNAs, and miRNAs were identified within each discordant pair and integrated with co-expression modules from external transcriptomic datasets constructed by Multiscale Embedded Gene Co-expression Network Analysis (MEGENA). This enabled the construction of disorder-specific ceRNA networks and the identification of core regulatory components.

We identified ceRNA networks for each disorder, revealing 19 miRNAs shared across all three disorders, while lncRNAs and mRNAs were primarily disorder-specific. Among the shared miRNAs, hsa-miR-29a-3p was downregulated in both SCZ and MDD, regulating distinct ceRNA axes involving COL6A6. Functional enrichment analysis of hub ceRNA networks revealed the convergent involvement of extracellular matrix (ECM)-receptor interaction pathways. Notably, COL6A6 (SCZ and MDD) and ITGB8 (BD) were key components of these pathways. Validation using independent brain and blood transcriptomic datasets demonstrated strong predictive potential for ITGB8 in the blood and prefrontal cortex for BD and SCZ, and moderate predictive potential for COL6A6 in the blood and anterior cingulate gyrus for SCZ.

This study identifies non-coding RNA–mediated regulatory networks implicated in the molecular etiology of psychiatric disorders. Our findings provide a foundation for precision diagnostics and targeted therapeutic strategies in psychiatry.

The online version contains supplementary material available at 10.1186/s12888-026-07894-5.