CRISPRLearner: A Deep Learning-Based System to Predict CRISPR/Cas9 sgRNA On-Target Cleavage Efficiency
Problem
Predicting the on-target cleavage efficiency of CRISPR/Cas9 sgRNA sequences so researchers can evaluate and redesign inefficient guides before experiments.
Why it matters
Poor guide design wastes costly gene-editing experiments and slows research across medicine and biotechnology.
What the AI did
A deep convolutional neural network automatically learns sequence determinants and predicts cleavage efficiency, usable with pre-trained models or custom-trained datasets, aided by a data augmentation technique.
Results
Average Spearman correlation higher than 0.40 across ten gene datasets/models
Impact
The tool provides a working efficiency predictor, though the reported correlation is modest.
Record
| Venue | Electronics |
|---|---|
| Published | 2019 |
| Work type | article |
| Authors | Giovanni Dimauro, Pierpasquale Colagrande, Roberto Carlucci, Mario Ventura, Vitoantonio Bevilacqua, Danilo Caivano |
| Citations | 39 (OpenAlex, at import time) |
| References | 28 |
| Model or method | deep convolutional neural network (with linear regression output) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.3390/electronics8121478 |
| Open access | gold |
| Keywords | CRISPR, sgRNA, deep learning, CNN, efficiency prediction |
Importance rating
real but narrower advance
Poor guide design wastes costly gene-editing experiments and slows research across medicine and biotechnology; abstract states no scale of impact.