Objective To investigate the role of mitochondrial autophagy-related genes in colorectal cancer using machine learning algorithms to analyze single-cell sequencing data and spatial transcriptomics data.

Methods Key mitochondrial autophagy-related genes were screened using Support Vector Machine-Recursive Feature Elimination (SVM-RFE) and Random Forest algorithms. Comprehensive assessments of cell-cell interactions, pseudo-time analysis, and gene spatial localization were conducted by integrating single-cell sequencing data and spatial transcriptomics data. Knockdown experiments were performed on the identified core genes to verify their effects on the proliferation, migration, and invasion capabilities of colorectal cancer cells.

Results RPS5 was identified as a core gene in the progression of colorectal cancer. RPS5-positive epithelial cells were primarily located in the tumor core region and interacted with fibroblasts. Immunohistochemical experiments demonstrated that RPS5 expression was significantly higher in colorectal cancer tissues compared to adjacent tissues. Functional experiments showed that knockdown of RPS5 significantly inhibited the proliferation, migration, and invasion capabilities of colorectal cancer cells.

Conclusion RPS5 is a key molecule promoting the progression of colorectal cancer, which provides a new potential target for precision therapy in colorectal cancer.