101 Machine Learning Algorithms for Mining Esophageal Squamous Cell Carcinoma Neoantigen Prognostic Models in Single-Cell Data.

ABSTRACT: Esophageal squamous cell carcinoma (ESCC) is one of the most aggressive malignant tumors in the digestive tract, characterized by a high recurrence rate and inadequate immunotherapy options. We analyzed mutation data of ESCC from public databases and employed 10 machine learning algorithms to generate 101 algorithm combinations. Based on the optimal range determined by the concordance index, we randomly selected one combination from the best-performing algorithms to construct a prognostic model consisting of five genes (DLX5, MAGEA4, PMEPA1, RCN1, and TIMP1). By validating the correlation between the prognostic model and antigen-presenting cells (APCs), we revealed the antigen-presentation efficacy of the model. Through the analysis of immune infiltration

SUBMITTER: Sun Y 

PROVIDER: S-EPMC11989522 | biostudies-literature | 2025 Apr

REPOSITORIES: biostudies-literature