2026-10-06T04:05:24Z https://api.figshare.com/v2/oai

oai:figshare.com:article/34073633 2026-10-05T17:49:59Z category_12 category_13 category_873 category_734 category_931 category_64 portal_5 item_type_3 month_year_10_2026

<p>GNNExplainer of genes.</p> Yue Yuan (113839) Yu Liu (6938) Jingchun Fan (9241232) Zhen Ren (2258311) Cell Biology Genetics Chemical Sciences not elsewhere classified Biological Sciences not elsewhere classified Information Systems not elsewhere classified Cancer virtual knockout techniques produced predicted probabilities like growth factor lethal malignancies worldwide enhance node features corresponding importance scores provided interpretable analyses fibronectin 1 (< colorectal &# 8220 cancer ” transition cancer &# 8221 differentially expressed genes graph level representations div >< p colorectal cancer groups key genes involved eight screened genes colorectal “ adenoma level representations colorectal cancer key genes enrichment analyses 33 (< genes affect stc2 </ sfrp1 </ sample classification precise identification overall survival nox4 </ named chebts multifeature learning mmp1 </ kyoto encyclopedia igfbp7 </ gse31905 datasets great significance geo database gene ontology following activation fn1 </ feature encoding extracellular mechanisms degs ). colorectal adenocarcinoma cldn2 </ biological processes <div><p>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 (<i>FN1</i>), Claudin 2 (<i>CLDN2</i>), Interleukin-33 (<i>IL-33</i>), Matrix Metallopeptidase 1 (<i>MMP1</i>), Stanniocalcin 2 (<i>STC2</i>), Insulin-Like Growth Factor-Binding Protein 7 (<i>IGFBP7</i>), NADPH Oxidase 4 (<i>NOX4</i>), and Secreted Frizzled Related Protein 1 (<i>SFRP1</i>), 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.</p></div> 2026-10-05T17:49:37Z Dataset Dataset 10.1371/journal.pone.0359785.t006 https://figshare.com/articles/dataset/_p_GNNExplainer_of_genes_p_/34073633 CC BY 4.0