Machine learning for identifying tumor stemness genes and developing prognostic model in gastric cancer

• DOI: 10.18632/aging.205715
• Publication Date: 2024-04-13T16:51:54Z
• Authors: Guo-Xing Li; Yun-Peng Chen; You-Yang Hu; Wen-Jing Zhao; Yun-Yan Lu; Fu-Jian Wan; Zhi-Jun Wu; Xiang-Qian Wang; Qi-Ying Yu
• Keywords: Tumor progression

Gastric cancer presents a formidable challenge, marked by its debilitating nature and often dire prognosis. Emerging evidence underscores the pivotal role of tumor stem cells in exacerbating treatment resistance fueling disease recurrence gastric cancer. Thus, identification genes contributing to stemness assumes paramount importance. Employing comprehensive approach encompassing ssGSEA, WGCNA, various machine learning algorithms, this study endeavors delineate key (TSKGs). Subsequently, these were harnessed construct prognostic model, termed Tumor Stemness Risk Genes Prognostic Model (TSRGPM). Through PCA, Cox regression analysis ROC curve analysis, efficacy Scores (TSRS) stratifying patient risk profiles was underscored, affirming ability as an independent indicator. Notably, TSRS exhibited significant correlation with lymph node metastasis Furthermore, leveraging algorithms such CIBERSORT dissect immune infiltration patterns revealed notable association between monocytes other cell. Subsequent scrutiny (TSRGs) culminated CDC25A for detailed investigation. Bioinformatics analyses unveil CDC25A's implication driving malignant phenotype tumors, discernible impact on cell proliferation DNA replication Noteworthy validation through vitro experiments corroborated bioinformatics findings, elucidating expression modulating In summation, established validated TSRGPM holds promise prognostication delineation potential therapeutic targets, thus heralding stride towards personalized management malignancy.