Deep Learning Identifies HAT1 as a Morphological Regulator in Esophageal Squamous Carcinoma Cells through Controlling Cell Senescence

Histopathology is a critical approach for diagnostic tasks and precision treatment. However, histopathological deep learning tools auto-identification remain poorly developed. Meanwhile, the interpretation of computer vision attention into cellular process less efficient in systematic way. Herein, it identified that histone acetyltransferase 1 (HAT1) an aging-associated gene esophagus epithelium by machine learning. An interpretable model developed to distinguish morphological changes with varied HAT1 expressions esophageal squamous carcinoma cells (ESCC). The gradient-weighted class activation mapping prediction score analysis reveal computer's focuses on nuclear sizes ESCC. hypothesized phenotype verified HAT1-knockdown ESCCs. Finally, regulating cell senescence affecting H3K27 acetylation E2F transcription factor 7 (E2F7) expression shown. feasibility benefits applying assistance systems routine practice scenarios connecting genotype further genetic research are suggested.