# Identification of ubiquitination-related signature genes for predicting kidney transplant rejection

## TL;DR

This study identifies six ubiquitination-related genes that can predict kidney transplant rejection and provides insights into immune mechanisms involved.

## Contribution

The study introduces a URGScore model and six signature genes for predicting kidney transplant rejection risk.

## Key findings

- Six signature genes (DTX3L, MARCH1, NCF4, RNF125, TRIM21, TRIM22) were identified for predicting kidney transplant rejection.
- The URGScore model achieved an AUC of 0.774 in predicting rejection risk in the discovery cohort.
- MARCH1 and RNF125 showed higher expression in transplant rejection cases compared to other biopsy samples.

## Abstract

Kidney transplant rejection (KTR) poses significant challenges to long-term graft survival, with involvement from ubiquitination-related genes (URGs) in immune modulation. This study aimed to identify key URGs linked to KTR and develop a predictive model for rejection risk. mRNA array data from the Gene Expression Omnibus were analyzed to find differentially expressed genes in GSE98320, which were intersected with URGs to yield 16 DE-URGs. Gene Ontology and KEGG enrichment analysis highlighted the NF-kappa B and TNF signaling pathways. A URGScore model stratified patients and revealed significant differences in immune cell infiltration, especially among Treg cells, demonstrating strong predictive performance in the discovery cohort (AUC = 0.774, 95% CI 0.747–0.800). Six signature genes (DTX3L, MARCH1, NCF4, RNF125, TRIM21, TRIM22) were identified, and their expression displayed a subtype-dependent gradient, increasing from antibody-mediated rejection to T cell-mediated rejection and reaching the highest levels in mixed rejection. These genes were incorporated into a nomogram, which achieved an AUC of 0.771 (95% CI 0.745–0.798). Validation in independent datasets confirmed the model’s reliability. In the two transplant rejection cases, MARCH1 and RNF125 showed higher expression than the other biopsy samples, while generalized high expression of all marker genes was observed in an IgA nephropathy patient. Together, these findings demonstrate the clinical relevance of URG-based biomarkers in KTR and provide molecular insight into immune-mediated rejection.

The online version contains supplementary material available at 10.1038/s41598-026-38022-8.