An overview and metanalysis of machine and deep learning-based CRISPR gRNA design tools
Problem
Predicting the on-target activity and off-target specificity of CRISPR-Cas9 guide RNAs is difficult, yet decisive for safe and effective gene editing.
Why it matters
gRNA activity and specificity determine the success and safety of genetic manipulation in academic and pharmaceutical applications.
What the AI did
The paper reviews machine and deep learning models and web tools that predict CRISPR gRNA on- and off-target activity from sequence features, making editing studies more convenient and precise.
Results
not stated in source
Impact
The review indicates MDL-based predictors are considered the most effective and reliable methods and are increasingly matching experimental observations.
Record
| Venue | RNA Biology |
|---|---|
| Published | 2019 |
| Work type | article |
| Authors | Jun Wang, Xiuqing Zhang, Lixin Cheng, Yonglun Luo |
| Citations | 102 (OpenAlex, at import time) |
| References | 104 |
| Model or method | various machine and deep learning models (reviewed) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1080/15476286.2019.1669406 |
| Open access | bronze |
| Keywords | CRISPR, gRNA design, machine learning, deep learning, off-target prediction |
Importance rating
significant — thousands to millions affected
Abstract states CRISPR-Cas9 is the most promising and versatile genetic manipulation tool, broadly adopted by academic and pharmaceutical communities; gRNA activity/specificity are decisive for any application, spanning a foundational field.