ClustalXeed: a GUI-based grid computation version for high performance and terabyte size multiple sequence alignment

BMC Bioinformatics

Software

Hyun Joo¹  Taeho Kim² 

[1] Department of Physiology and Integrated Biosystems, College of Medicine, Inje University, 614-735, Busan, South Korea;Laboratory of Systems Immunology, World Premier International Immunology Frontier Research Center, Osaka University, 565-0871, Suita, Osaka, Japan;

关键词: Multiple Sequence Alignment;    Message Passing Interface;    Random Access Memory;    Slave Node;    Computation Node;   

DOI  :  10.1186/1471-2105-11-467

 received in 2010-05-06, accepted in 2010-09-17,  发布年份 2010

来源: Springer

【 摘 要 】

BackgroundThere is an increasing demand to assemble and align large-scale biological sequence data sets. The commonly used multiple sequence alignment programs are still limited in their ability to handle very large amounts of sequences because the system lacks a scalable high-performance computing (HPC) environment with a greatly extended data storage capacity.ResultsWe designed ClustalXeed, a software system for multiple sequence alignment with incremental improvements over previous versions of the ClustalX and ClustalW-MPI software. The primary advantage of ClustalXeed over other multiple sequence alignment software is its ability to align a large family of protein or nucleic acid sequences. To solve the conventional memory-dependency problem, ClustalXeed uses both physical random access memory (RAM) and a distributed file-allocation system for distance matrix construction and pair-align computation. The computation efficiency of disk-storage system was markedly improved by implementing an efficient load-balancing algorithm, called "idle node-seeking task algorithm" (INSTA). The new editing option and the graphical user interface (GUI) provide ready access to a parallel-computing environment for users who seek fast and easy alignment of large DNA and protein sequence sets.ConclusionsClustalXeed can now compute a large volume of biological sequence data sets, which were not tractable in any other parallel or single MSA program. The main developments include: 1) the ability to tackle larger sequence alignment problems than possible with previous systems through markedly improved storage-handling capabilities. 2) Implementing an efficient task load-balancing algorithm, INSTA, which improves overall processing times for multiple sequence alignment with input sequences of non-uniform length. 3) Support for both single PC and distributed cluster systems.

【 授权许可】

CC BY   
© Kim and Joo; licensee BioMed Central Ltd. 2010