RNA-targeting
RNA-targeting predicts Cas13d (CasRx/RfxCas13d) guide RNA efficiency and evaluates Cas13d orthologs to enable precise RNA perturbation with reduced cellular toxicity in mammalian systems.
Key Features:
- Guide Efficiency Prediction: Predicts CasRx (RfxCas13d) guide RNA efficiency across all human and mouse RefSeq genes based on an analysis of over 127,000 guide RNAs.
- Machine Learning Models: Evaluates three linear models, two ensemble models, and two deep learning models for guide efficiency prediction.
- Deep Learning Insights: The deep learning model identifies specific sequence motifs and secondary features that contribute to high-efficiency guides at spacer positions 15-24.
- Orthogonal Validation: Validates the prediction algorithm across multiple human cell types using orthogonal experiments.
- Custom Sequence Input: Accepts custom target sequences for CasRx guide design.
- Novel Ortholog Discovery: Incorporates metagenomic mining results that identified 46 novel Cas13d orthologs, including DjCas13d with low cellular toxicity and high specificity in human embryonic stem cells, neural progenitor cells, and neurons.
- Generalization Across Orthologs: Generalizes the CasRx-trained guide efficiency model to the DjCas13d ortholog.
Scientific Applications:
- Transcriptome engineering: Enables precise RNA knockdown for transcriptome engineering with reduced off-target effects.
- Functional genomics: Facilitates discovery of gene function through targeted transcript perturbation.
- Disease modeling: Supports modeling diseases by enabling specific RNA perturbations.
- RNA therapeutics development: Assists design of high-efficiency, low-toxicity guide RNAs for therapeutic applications.
Methodology:
Evaluated three linear models, two ensemble models, and two deep learning models on a dataset of over 127,000 guide RNAs; the deep learning model identified sequence motifs and secondary features at spacer positions 15-24; metagenomic mining identified 46 novel Cas13d orthologs; the prediction algorithm was validated via orthogonal experiments across multiple human cell types.
Topics
Details
- License:
- Not licensed
- Cost:
- Free of charge
- Tool Type:
- web application
- Operating Systems:
- Mac, Linux, Windows
- Added:
- 2/3/2022
- Last Updated:
- 2/3/2022
Operations
Publications
Wei J, Lotfy P, Faizi K, Baungaard S, Gibson E, Wang E, Slabodkin H, Kinnaman E, Chandrasekaran S, Kitano H, Durrant MG, Duffy CV, Hsu PD, Konermann S. Deep learning and CRISPR-Cas13d ortholog discovery for optimized RNA targeting. Unknown Journal. 2021. doi:10.1101/2021.09.14.460134.