Abstract | ||
---|---|---|
Simulated annealing (SA) is one of the popular approaches to predict protein structures. SA is prohibitive because it usually consumes much computing time and is likely to fall into local minimum points. We proposed a parallel SA algorithm based on a Graph Process Unit (GPU) technique to improve the efficiency and accuracy of the protein structure prediction. First, we analyze the SA algorithm based on CPU, second, we introduce the architecture of Compute Unified Device Architecture (CUDA). Finally, we applied statistical method to optimize the performance of the CUDA based parallel algorithm. The experimental result shows that our algorithm provides a feasible solution for the protein structure prediction. |
Year | DOI | Venue |
---|---|---|
2012 | 10.1109/ICMLA.2012.117 | ICMLA (1) |
Keywords | Field | DocType |
simulated annealing,gpu,prediction of protein structures,statistical analysis,parallel architectures,sa algorithm,graph process unit technique,graphics processing units,sa,proteins,compute unified device architecture,cuda,biology computing,computing time,gpu based simulated annealing,parallel sa algorithm,local minimum points,parallel algorithm,protein structure,cpu,protein structure prediction,graph process unit,statistical method | Simulated annealing,Graph,Protein structure prediction,Computer science,Parallel algorithm,CUDA,Adaptive simulated annealing,Computational science,Artificial intelligence,Machine learning,Protein structure,Statistical analysis | Conference |
Volume | ISBN | Citations |
1 | 978-1-4673-4651-1 | 1 |
PageRank | References | Authors |
0.34 | 5 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Hui Li | 1 | 3 | 5.46 |
Chun-Mei Liu | 2 | 245 | 41.30 |