Related Experiment Video
Updated: Jun 1, 2026
09:34
A Virtual Machine Platform for Non-Computer Professionals for Using Deep Learning to Classify Biological Sequences of Metagenomic Data
Published on: September 25, 2021
DeepCas12a: a hybrid deep learning framework for accurate AsCas12a efficiency prediction from sequence and epigenetic
Yiming Shi1,2, Junkai Yin1,2, Shurui Ning1,2
1Department of Infectious Dermatosis, Center of Infectious Skin Diseases, Bioinformatics Department, School of Life Sciences and Technology, Shanghai Skin Disease Hospital, Tongji University, Shanghai, 200092, China.
BMC Genomics
|May 30, 2026
Summary
DeepCas12a, a new deep learning tool, improves CRISPR-Cas12a genome editing efficiency prediction by analyzing DNA sequences and epigenetic data. It outperforms existing methods for guide RNA design.
Area of Science:
- Genomics
- Bioinformatics
- Molecular Biology
Background:
- CRISPR-Cas12a (Cpf1) is a versatile genome editing tool with T-rich PAM recognition.
- Predicting Cas12a cleavage efficiency is challenging due to sequence context and epigenetic factors influencing high-order interactions.
Purpose of the Study:
- To develop an advanced deep learning framework, DeepCas12a, for accurate prediction of CRISPR-Cas12a cleavage efficiency.
- To integrate multimodal data, including DNA sequences and epigenetic profiles, for enhanced predictive performance.
Main Methods:
- A hybrid deep learning model combining Convolutional Neural Networks (CNNs) and a Vision Transformer (ViT) encoder.
- End-to-end architecture fusing DNA sequence data with epigenetic profiles (DNA methylation, chromatin accessibility).
Main Results:
- DeepCas12a achieved superior performance on an independent test set, with an Average Precision of 0.783, AUC of 0.868, and Spearman correlation of 0.630.
- Interpretability analysis using saliency maps confirmed the model's ability to identify biologically relevant features like PAM specificity and seed region sensitivity.
Conclusions:
- DeepCas12a offers a significant advancement in predicting CRISPR-Cas12a activity.
- The model facilitates rational guide RNA design by capturing complex sequence and epigenetic interactions.