Related Experiment Video
Updated: Oct 29, 2025

11:14
Detection of Low Copy Number Integrated Viral DNA Formed by In Vitro Hepatitis B Infection
Published on: November 7, 2018
18.6K
DeepHBV: a deep learning model to predict hepatitis B virus (HBV) integration sites.
Canbiao Wu1, Xiaofang Guo2, Mengyuan Li3
1Institute for Brain Research and Rehabilitation, South China Normal University, Guangzhou, 510631, Guangdong, China.
BMC Ecology and Evolution
|July 8, 2021
Summary
A new deep learning model, DeepHBV, accurately predicts hepatitis B virus (HBV) integration sites. This tool enhances understanding of HBV-related liver cancer by identifying key genomic features and transcription factor binding sites.
Area of Science:
- Genomics
- Bioinformatics
- Computational Biology
Background:
- Hepatitis B virus (HBV) is a primary cause of viral hepatitis and liver cancer.
- HBV integration into the host genome is a critical step in virus-induced malignant transformation.
Purpose of the Study:
- To develop and validate DeepHBV, an attention-based deep learning model for predicting HBV integration sites.
- To identify genomic features and transcription factor binding sites associated with HBV integration.
Main Methods:
- Developed an attention-based deep learning model, DeepHBV.
- Trained and tested DeepHBV using HBV integration site data from the dsVIS database.
- Integrated genomic features, including repeat peaks and TCGA Pan-Cancer peaks, to enhance model performance.
Main Results:
- DeepHBV achieved significant improvements in prediction accuracy with integrated genomic features (AUROC up to 0.9430, AUPR up to 0.9310).
- Identified enrichment of specific transcription factor binding sites (TFBS) near HBV integration sites.
- Highlighted binding sites for factors including AR-halfsite, Arnt, Atf1, bHLHE40, bHLHE41, BMAL1, CLOCK, c-Myc, COUP-TFII, E2A, EBF1, Erra, and Foxo3.
Conclusions:
- DeepHBV serves as a valuable tool for predicting HBV integration sites.
- The study provides novel insights into HBV integration preferences and their role in carcinogenesis.

