We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
CURC: a CUDA-based reference-free read compressor.
- Authors
Xie, Shaohui; He, Xiaotian; He, Shan; Zhu, Zexuan
- Abstract
Motivation The data deluge of high-throughput sequencing (HTS) has posed great challenges to data storage and transfer. Many specific compression tools have been developed to solve this problem. However, most of the existing compressors are based on central processing unit (CPU) platform, which might be inefficient and expensive to handle large-scale HTS data. With the popularization of graphics processing units (GPUs), GPU-compatible sequencing data compressors become desirable to exploit the computing power of GPUs. Results We present a GPU-accelerated reference-free read compressor, namely CURC, for FASTQ files. Under a GPU–CPU heterogeneous parallel scheme, CURC implements highly efficient lossless compression of DNA stream based on the pseudogenome approach and CUDA library. CURC achieves 2–6-fold speedup of the compression with competitive compression rate, compared with other state-of-the-art reference-free read compressors. Availability and implementation CURC can be downloaded from https://github.com/BioinfoSZU/CURC. Supplementary information Supplementary data are available at Bioinformatics online.
- Subjects
INTERNET servers; CENTRAL processing units; GRAPHICS processing units; COMPRESSORS; DATA warehousing; NUCLEOTIDE sequencing
- Publication
Bioinformatics, 2022, Vol 38, Issue 12, p3294
- ISSN
1367-4803
- Publication type
Article
- DOI
10.1093/bioinformatics/btac333