QuickProbs
QuickProbs performs multiple sequence alignment by implementing GPU-optimized versions of the MSAProbs algorithm to accelerate alignment while preserving or improving accuracy for applications such as secondary structure prediction, domain searching, and phylogenetic analysis.
Key Features:
- GPU Optimization: Uses OpenCL to execute alignment computations on graphics processing units, enabling compatibility across GPU vendors.
- Targeted Redesign: Redesigns the calculation of posterior matrices and the consistency transformation from MSAProbs for execution on GPUs.
- Performance Gains: Demonstrates 5.7–9.7× speed-up over CPU-parallel MSAProbs on BAliBASE, PREFAB, and OXBench-X benchmarks on a quad-core PC with a high-end graphics card, and an overall 6.7× speed-up on Pfam protein families.
- Accuracy: Maintains or surpasses the accuracy of MSAProbs and is reported to be more accurate than MAFFT, MUSCLE, and ClustalW at comparable speeds.
- Enhanced Variant: Provides a tuned variant that improves accuracy for sets of distantly related sequences without exceeding the computational time of MSAProbs.
Scientific Applications:
- Secondary Structure Prediction: Supplies accurate multiple sequence alignments to support protein secondary structure prediction.
- Domain Searching: Identifies conserved regions across sequences to facilitate domain searches.
- Phylogenetic Analysis: Produces alignments suitable for constructing phylogenetic trees and studying evolutionary relationships in large datasets.
Methodology:
Builds upon the MSAProbs algorithm by executing the posterior matrix calculation and consistency transformation on GPUs via OpenCL.
Topics
Details
- Tool Type:
- command-line tool
- Operating Systems:
- Linux, Windows
- Programming Languages:
- C++, Perl, C
- Added:
- 8/3/2017
- Last Updated:
- 11/25/2024
Operations
Publications
Gudyś A, Deorowicz S. QuickProbs—A Fast Multiple Sequence Alignment Algorithm Designed for Graphics Processors. PLoS ONE. 2014;9(2):e88901. doi:10.1371/journal.pone.0088901. PMID:24586435. PMCID:PMC3934876.