Multi-GPU Approach for Large-Scale Multiple Sequence Alignment

Abstract Motivation The success of genome sequencing techniques has resulted in rapid explosion of protein sequences. Collections of multiple homologous sequences can provide critical information to the modeling of structure and function of unknown proteins. There are however no standard and efficient pipeline available for sensitive multiple sequence alignment (MSA) collection. This is particularly challenging when large whole-genome and metagenome databases are involved. Results We developed DeepMSA, a new open-source method for sensitive MSA construction, which has homologous sequences and alignments created from multi-sources of whole-genome and metagenome databases through complementary hidden Markov model algorithms. The practical usefulness of the pipeline was examined in three large-scale benchmark experiments based on 614 non-redundant proteins. First, DeepMSA was utilized to generate MSAs for residue-level contact prediction by six coevolution and deep learning-based programs, which resulted in an accuracy increase in long-range contacts by up to 24.4% compared to the default programs. Next, multiple threading programs are performed for homologous structure identification, where the average TM-score of the template alignments has over 7.5% increases with the use of the new DeepMSA profiles. Finally, DeepMSA was used for secondary structure prediction and resulted in statistically significant improvements in the Q3 accuracy. It is noted that all these improvements were achieved without re-training the parameters and neural-network models, demonstrating the robustness and general usefulness of the DeepMSA in protein structural bioinformatics applications, especially for targets without homologous templates in the PDB library. Availability and implementation https://zhanglab.ccmb.med.umich.edu/DeepMSA/. Supplementary information Supplementary data are available at Bioinformatics online.

Download Full-text

A multiple sequence alignment method with sequence vectorization

Engineering Computations ◽

10.1108/ec-01-2013-0026 ◽

2014 ◽

Vol 31 (2) ◽

pp. 283-296

Author(s):

Guoli Ji ◽

Yong Zeng ◽

Zijiang Yang ◽

Congting Ye ◽

Jingci Yao

Keyword(s):

Sequence Alignment ◽

Multiple Sequence Alignment ◽

Time Complexity ◽

Large Scale ◽

Distance Matrix ◽

Traditional Methods ◽

Multiple Sequence ◽

Guide Tree ◽

Content Type ◽

Matrix Calculation

Purpose – The time complexity of most multiple sequence alignment algorithm is O(N2) or O(N3) (N is the number of sequences). In addition, with the development of biotechnology, the amount of biological sequences grows significantly. The traditional methods have some difficulties in handling large-scale sequence. The proposed Lemk_MSA method aims to reduce the time complexity, especially for large-scale sequences. At the same time, it can keep similar accuracy level compared to the traditional methods. Design/methodology/approach – LemK_MSA converts multiple sequence alignment into corresponding 10D vector alignment by ten types of copy modes based on Lempel-Ziv. Then, it uses k-means algorithm and NJ algorithm to divide the sequences into several groups and calculate guide tree of each group. A complete guide tree for multiple sequence alignment could be constructed by merging guide tree of every group. Moreover, for large-scale multiple sequence, Lemk_MSA proposes a GPU-based parallel way for distance matrix calculation. Findings – Under this approach, the time efficiency to process multiple sequence alignment can be improved. The high-throughput mouse antibody sequences are used to validate the proposed method. Compared to ClustalW, MAFFT and Mbed, LemK_MSA is more than ten times efficient while ensuring the alignment accuracy at the same time. Originality/value – This paper proposes a novel method with sequence vectorization for multiple sequence alignment based on Lempel-Ziv. A GPU-based parallel method has been designed for large-scale distance matrix calculation. It provides a new way for multiple sequence alignment research.

Download Full-text