CRISPRon

CRISPRon predicts CRISPR-Cas9 guide RNA (gRNA) on-target efficiency using a deep learning model trained on experimentally generated and published gRNA activity datasets.

Key Features:

• Deep learning model: Employs a deep learning algorithm to predict SpCas9 gRNA on-target efficiency.
• Training dataset: Trained on a combined dataset of 23,902 gRNAs, including 10,592 experimentally generated SpCas9 gRNA activity measurements and complementary published datasets.
• High-quality activity data: Incorporates high-quality on-target gRNA activity measurements to inform model training.
• Validation: Evaluated on four independent test datasets not used in training, showing superior predictive performance compared to existing tools.
• Improved accuracy: Delivers significantly improved prediction performance for gRNA on-target efficiency relative to prior methods.

Scientific Applications:

• gRNA selection: Prioritizes high-efficiency gRNAs for CRISPR-Cas9 experiments to maximize on-target activity.
• Genome editing optimization: Improves success rates of genome editing projects by providing more reliable on-target efficiency predictions.
• Experimental design support: Informs design of genetic research and biotechnology experiments by ranking candidate guides.

Methodology:

Deep learning model trained on a combined dataset of 23,902 gRNAs (10,592 experimentally generated SpCas9 gRNAs plus complementary published datasets) and evaluated on four independent test datasets.