Prediction of Cell States and Key Transcription Factors of the Human Cornea Through Integrated Single-Cell Omics Analyses

This study integrates single-cell RNA sequencing data from four studies to create a comprehensive meta-atlas of human corneal cell states, identifying 21 distinct cell states, including novel markers and rare cell types like nonmyelinating corneal Schwann cells. A machine learning tool, cPredictor, was developed to accurately annotate these cell states, achieving a high weighted F1 score of 95.75% in cross-validation. The study also identifies key transcription factors, such as TP63, FOSL2, and PAX6, crucial for corneal cell state determination. The findings highlight differences between iPSC-derived corneal organoids and adult corneas, suggesting the need for improved maturation techniques. This work provides a robust reference for future research in corneal development, disease, and regeneration.