Related Experiment Video
Updated: Jun 22, 2025
10:46
Dissection of Enhancer Function Using Multiplex CRISPR-based Enhancer Interference in Cell Lines
Published on: June 2, 2018
9.3K
CapsEnhancer: An Effective Computational Framework for Identifying Enhancers Based on Chaos Game Representation and
Lantian Yao1,2, Peilin Xie1, Jiahui Guan3
1Kobilka Institute of Innovative Drug Discovery, School of Medicine, The Chinese University of Hong Kong, Shenzhen 518172, China.
Journal of Chemical Information and Modeling
|July 1, 2024
Summary
A new deep learning method, CapsEnhancer, accurately predicts gene enhancers and their strengths using DNA sequence images. This computational approach offers a resource-efficient alternative to traditional experimental methods for enhancer identification.
Area of Science:
- Genomics
- Bioinformatics
- Computational Biology
Background:
- Enhancers are critical noncoding DNA regulatory elements controlling gene expression.
- Experimental enhancer identification is resource-intensive.
- Computational prediction of enhancers is increasingly important.
Purpose of the Study:
- To develop a deep learning framework for accurate enhancer identification and strength prediction.
- To introduce a novel computational method for analyzing biological sequences.
Main Methods:
- A two-stage deep learning framework, CapsEnhancer, was developed.
- Chaos game representation encoded DNA sequences into images.
- A capsule network was used for feature extraction from sequence images.
Main Results:
- CapsEnhancer achieved state-of-the-art performance in enhancer identification and strength prediction.
- Accuracy reached 94.5% in the first stage and 95% in the second stage, surpassing previous methods.
- This study pioneered the use of computer vision in enhancer identification.
Conclusions:
- CapsEnhancer provides an accurate and efficient computational method for enhancer identification.
- The approach offers a novel perspective for biological sequence analysis.
- This work advances the understanding of gene regulation through enhancer prediction.
