Exploring the key genes of colorectal “adenoma-cancer” based on graph transformer
- 1School of Information Engineering, Gansu University of Chinese Medicine, Lanzhou, China
- 2Center for Evidence-based Medicine, Gansu University of Chinese Medicine, Lanzhou, China
Abstract
Colorectal cancer (CRC) is one of the most lethal malignancies worldwide, and the precise identification of biomarkers from colonic adenoma to cancer is of great significance for preventing the development of adenocarcinoma. Given that existing methods inadequately capture the topological network relationships among genes, this study proposes a graph neural network model based on multifeature learning, named ChebTs, to investigate the correlation between key genes involved in the colorectal “adenoma-cancer” transition. The GSE41657 and GSE31905 datasets from the GEO database were stratified into normal, adenoma, and colorectal cancer groups. Feature encoding was introduced to enhance node features, followed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of differentially expressed genes (DEGs). A protein-protein interaction (PPI) network was constructed using Cytoscape software, and the aggregated information was embedded into the model for training to generate a list of key genes with corresponding importance scores. An attention pooling mechanism aggregated node-level representations into graph level representations for sample classification, and a two layer fully connected network, following activation and regularization, produced predicted probabilities. Furthermore, we provided interpretable analyses at the gene level using GNNExplainer. The results were validated through virtual knockout techniques. The eight screened genes, Fibronectin 1 (FN1), Claudin 2 (CLDN2), Interleukin-33 (IL-33), Matrix Metallopeptidase 1 (MMP1), Stanniocalcin 2 (STC2), Insulin-Like Growth Factor-Binding Protein 7 (IGFBP7), NADPH Oxidase 4 (NOX4), and Secreted Frizzled Related Protein 1 (SFRP1), were all found to be associated with overall survival (OS) in CRC. In this study, eight molecules closely related to the development of colorectal adenocarcinoma were screened out, and they may be diagnostic biomarkers of colorectal cancer. These genes affect the prognosis of patients by participating in biological processes such as remodeling of extracellular mechanisms, and are of great significance for preventing the carcinogenesis of adenoma.
Related
- Microbial Invasion and Immunosuppression Drives Adenoma Progression in Early Colorectal Cancer Development
- Metabolic syndrome and colorectal cancer share GPAT3 and PLAU as candidate genes: Mendelian randomisation, machine learning, single-cell knockout and docking
- Network pharmacology and single-cell sequencing analysis of gastrodin in anticancer effects against colorectal cancer by regulating key target genes and the tumor immune microenvironment.
- Tumor-to-tumor metastasis of colorectal adenocarcinoma within a gonadotroph pituitary adenoma.
- Effect modification of gallbladder wall thickness on the association between fibrosis-4 index and high-risk colorectal adenoma in patients with gallbladder polyps: a retrospective cross-sectional study.
BibTeX
@article{yuan2026exploring,
title = {{Exploring the key genes of colorectal “adenoma-cancer” based on graph transformer}},
author = {Yue Yuan and Yu Liu and Jingchun Fan and Zhen Ren},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0359785},
url = {https://doi.org/10.1371/journal.pone.0359785},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Exploring the key genes of colorectal “adenoma-cancer” based on graph transformer
AU - Yue Yuan
AU - Yu Liu
AU - Jingchun Fan
AU - Zhen Ren
PY - 2026
DA - 2026/10/05
JO - PLOS One
DO - 10.1371/journal.pone.0359785
UR - https://doi.org/10.1371/journal.pone.0359785
AB - Colorectal cancer (CRC) is one of the most lethal malignancies worldwide, and the precise identification of biomarkers from colonic adenoma to cancer is of great significance for preventing the development of adenocarcinoma. Given that existing methods inadequately capture the topological network relationships among genes, this study proposes a graph neural network model based on multifeature learning, named ChebTs, to investigate the correlation between key genes involved in the colorectal “adenoma-cancer” transition. The GSE41657 and GSE31905 datasets from the GEO database were stratified into normal, adenoma, and colorectal cancer groups. Feature encoding was introduced to enhance node features, followed by Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of differentially expressed genes (DEGs). A protein-protein interaction (PPI) network was constructed using Cytoscape software, and the aggregated information was embedded into the model for training to generate a list of key genes with corresponding importance scores. An attention pooling mechanism aggregated node-level representations into graph level representations for sample classification, and a two layer fully connected network, following activation and regularization, produced predicted probabilities. Furthermore, we provided interpretable analyses at the gene level using GNNExplainer. The results were validated through virtual knockout techniques. The eight screened genes, Fibronectin 1 (FN1), Claudin 2 (CLDN2), Interleukin-33 (IL-33), Matrix Metallopeptidase 1 (MMP1), Stanniocalcin 2 (STC2), Insulin-Like Growth Factor-Binding Protein 7 (IGFBP7), NADPH Oxidase 4 (NOX4), and Secreted Frizzled Related Protein 1 (SFRP1), were all found to be associated with overall survival (OS) in CRC. In this study, eight molecules closely related to the development of colorectal adenocarcinoma were screened out, and they may be diagnostic biomarkers of colorectal cancer. These genes affect the prognosis of patients by participating in biological processes such as remodeling of extracellular mechanisms, and are of great significance for preventing the carcinogenesis of adenoma.
SN - 1932-6203
ER -