Macrogenome-based method for multiple-sequence alignment of proteins

Abstract

The invention relates to a macrogenome-based method for multiple-sequence alignment of proteins, which comprises the following steps: performing initial search on a UniClust30 database by using HHblits according to a sequence of a target protein, filtering out a sequence with residue gap exceeding 50% the target sequence length in searched MSA, and iterating the search process for three times; filtering obtained multi-sequence comparison files by using hhfilter to generate a sequence with a gap proportion exceeding 25% in the MSA, so as to obtain MSA1; calculating the valid sequence number Meff of MSA1, if Meff is larger than or equal to 10 L, ending the search, and taking the MSA1 as an output result of multi-sequence alignment; otherwise, constructing a hidden Markov model HMM of the MSA1, searching a Metaclust50 metagenome database to obtain multiple-sequence alignment MSA2, and combining the MSA1 with the MSA2 to obtain final MSA. The invention not only enhances the number and quality of the searched homologous sequences, but also improves the calculation efficiency.

Description

technical field [0001] The invention relates to the fields of bioinformatics and computer applications, in particular to a metagenomic-based protein multiple sequence alignment method. Background technique [0002] Proteins are the focus of many areas of life science research because they are responsible for most of the biological functions of living organisms. Sequence alignment is the basic composition and important foundation of bioinformatics. The basic idea of ​​sequence alignment is that, based on the general law that sequence determines structure and structure determines function in biology, both nucleic acid sequences and sequences on the primary structure of proteins are regarded as strings composed of basic characters, and the differences between sequences are detected. Similarity, discovering information about function, structure and evolution in biological sequences. The theoretical basis of sequence comparison is the theory of evolution. If there is enough sim...

AI Technical Summary

Problems solved by technology

However, with many target protein sequences, especially novel proteins, it is difficult to search sufficient homologous sequences in databases to reliably extract evolutionary information to guide protein folding

[0004] In summary, the existing protein multiple sequence alignment methods are not perfect in terms of the quantity and quality of the search for homologous sequences and computational efficiency, resulting in the inability to reliably extract evolutionary information and then guide protein folding, so improvements are needed

Images