MUMmerGPU: Fast Sequence Alignment Using GPUs
Author Information
Author(s): Michael C Schatz, Cole Trapnell, Arthur L Delcher, Amitabh Varshney
Primary Institution: Center for Bioinformatics and Computational Biology, University of Maryland, College Park, MD, USA
Hypothesis
Can high-throughput sequence alignment be significantly accelerated using Graphics Processing Units (GPUs)?
Conclusion
MUMmerGPU is a low-cost, ultra-fast sequence alignment program that outperforms traditional CPU-based methods.
Supporting Evidence
- MUMmerGPU achieves more than a 10-fold speedup over a serial CPU version of the sequence alignment kernel.
- It outperforms the exact alignment component of MUMmer by 3.5-fold in total application time.
- MUMmerGPU can handle the increasing volume of data produced by new, high-throughput sequencing technologies.
Takeaway
MUMmerGPU is a program that helps scientists quickly compare DNA sequences using powerful graphics cards, making it much faster than older methods.
Methodology
MUMmerGPU uses the Compute Unified Device Architecture (CUDA) to perform parallel pairwise local sequence alignment on GPUs.
Limitations
The performance may decrease for longer reads due to poor cache performance and thread divergence.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website