Related Experiment Video
Updated: Mar 17, 2026
11:14
Detection of Low Copy Number Integrated Viral DNA Formed by In Vitro Hepatitis B Infection
Published on: November 7, 2018
18.9K
DeF-GPU: Efficient and effective deletions finding in hepatitis B viral genomic DNA using a GPU architecture
Chun-Pei Cheng1, Kuo-Lun Lan1, Wen-Chun Liu2
1Department of Computer Science and Information Engineering, National Cheng Kung University, Tainan 701, Taiwan.
Methods (San Diego, Calif.)
|August 3, 2016
Summary
Identifying Hepatitis B virus (HBV) genomic deletions is crucial for understanding liver disease. A new GPU-based method, DeF-GPU, efficiently and accurately detects these deletions in large next-generation sequencing datasets.
Area of Science:
- Virology
- Bioinformatics
- Genomics
Background:
- Hepatitis B virus (HBV) infection increases the risk of liver diseases, including cirrhosis and hepatocellular carcinoma (HCC).
- Genomic DNA deletions in HBV are linked to viral activity and host immune responses.
- Identifying HBV deletions is challenging due to large and complex biological datasets.
Purpose of the Study:
- To develop an efficient and precise method for identifying HBV genomic deletions from next-generation sequencing (NGS) data.
- To address the limitations of existing tools not optimized for viral sequence analysis.
Main Methods:
- Proposed a graphics processing unit (GPU)-based data mining method named DeF-GPU.
- Utilized Compute Unified Device Architecture (CUDA) for parallelizing deletion identification procedures.
- Designed the method to handle large NGS datasets containing millions of reads.
Main Results:
- DeF-GPU demonstrated superior performance compared to the commonly used Pindel method.
- The method accurately identified ground truth deletions in synthetic and real datasets.
- DeF-GPU achieved high precision and efficiency, identifying deletions in seconds.
Conclusions:
- DeF-GPU offers an efficient and accurate solution for HBV deletion detection in large NGS datasets.
- The developed method can aid in understanding HBV-associated liver diseases.
- The source code is publicly available for further research and application.
