High-Throughput Validation of Cis-Regulatory Elements in Maize Using UMI-STARR-seq

This project demonstrates the genome-wide identification and high-throughput validation of cis-regulatory elements (CREs) in maize. By employing sequence-to-expression deep learning models, we predicted CREs across the maize genome and identified key distal and proximal elements. To validate these predictions, 12,000 sequences were synthesized and tested for activity using UMI-STARR-seq in maize protoplasts. This integrated approach, combining computational prediction with experimental validation, provides a comprehensive framework for understanding gene regulation in maize and paves the way for the precise engineering of gene expression to improve crop traits.