CN106446604A - Protein structure ab into prediction method based on firefly algorithm - Google Patents

Abstract

The invention discloses a protein structure ab into prediction method based on a firefly algorithm. The method includes that under a basic firefly algorithm frame, a coarseness energy model is adopted to effectively lower conformational space dimension, group property of the firefly algorithm is utilized to guarantee diversity of protein conformation, segment assembling technology is adopted to initialize conformational group, a dihedral angle is used to express position of conformation in space according to a coarseness expression model of the protein conformation, energy ranking is adopted to determine a strongest luminous individual, position of the conformation is updated by calculating attraction degree among individuals, and approximately-natural-state conformation with lowest energy is acquired by searching in the conformational space. By applying the method in protein structure prediction, conformation high in predication accuracy and low in complexity can be acquired.

Description

An Ab Initio Method for Protein Structure Prediction Based on the Firefly Algorithm

technical field

The invention relates to the fields of bioinformatics and computer applications, and in particular to a method for protein structure prediction from scratch based on firefly algorithm.

Background technique

Bioinformatics is a research hotspot in the intersection of life science and computer science. Bioinformatics research results have been widely used in gene discovery and prediction, gene data storage and management, data retrieval and mining, gene expression data analysis, protein structure prediction, gene and protein homology relationship prediction, sequence analysis and comparison, etc. . The genome specifies all the proteins that make up the organism, and the genes specify the sequence of amino acids that make up the proteins. Although proteins are composed of linear sequences of amino acids, they can only have corresponding activities and corresponding biological functions if they are folded to form a specific spatial structure. Knowing the spatial structure of proteins is not only conducive to understanding the functions of proteins, but also helps to understand how proteins perform their functions. Determining the structure of a protein is very important. At present, the data accumulation speed of the protein sequence database is very fast, but there are relatively few proteins with known structures. Although protein structure determination technology has made significant progress, the process of determining protein structure through experimental methods is still very complicated and expensive. Consequently, experimentally determined protein structures are far fewer than known protein sequences. On the other hand, with the development of DNA sequencing technology, the human genome and more model organism genomes have been or will be completely sequenced, and the number of DNA sequences will increase rapidly. Due to the advancement of DNA sequence analysis technology and gene identification methods, we A large number of protein sequences can be deduced from DNA. This means that the gap between the number of proteins with known sequences and the number of proteins with determined structures (such as data in the protein structure database PDB) will become larger and larger. It is hoped that the rate at which protein structures can be produced can keep up with the rate at which protein sequences can be produced, or close the gap between the two.

At present, the main technical bottleneck lies in two aspects. The first aspect lies in the sampling method. The existing technology is not strong enough to sample the conformation space. Therefore, existing conformational space search methods are deficient and need to be improved.

Contents of the invention

In order to overcome the disadvantages of low sampling efficiency, high complexity, and low prediction accuracy in existing methods for protein structure prediction and conformation space optimization, the present invention proposes an ab initio method for protein structure prediction based on the firefly algorithm. Under the framework of the basic firefly algorithm, the coarse-grained energy model is used to effectively reduce the dimension of the conformational space, the population characteristics of the firefly algorithm are used to ensure the diversity of protein conformations, and the fragment assembly technology is used to initialize the conformational population. Express the model, use a set of dihedral angles to represent the position of the conformation in space, use energy ranking to determine the strongest luminescent individual, and update the position of the conformation by calculating the attraction between individuals, and finally search for the minimum energy in the conformation space near-native conformation.

The technical solution adopted by the present invention to solve its technical problems is:

A de novo method for protein structure prediction based on the firefly algorithm, the method comprising the following steps:

1) given input sequence information;

2) Parameter initialization: set the population size popSize, the number of iterations generation, the light intensity attraction factor γ, and the position update step factor α;

3) Population conformation initialization: According to the given input sequence, randomly generate popSize individuals, perform length fragment assembly for each individual in the population, and calculate its fluorescence brightness Io, where length is the sequence length, Io=-E, E is the protein conformational energy value calculated by the RosettaSscore3 energy function;

4) sort the fluorescence brightness calculated in step 3) from large to small, so that the individual with the largest fluorescence brightness is p _g ;

5) Start iteration:

5.1) For each individual in the group, calculate the attractiveness β of p _g to it;

5.2) Update the position of each individual in the space according to x _i (t+1)= _xi (t)+β(x _j (t) _–xi (t))+α(rand–0.5), where x _i (t+1), x _i (t) represents the updated position and current position of individual p _i , x _j (t) represents the current position of individual p _g , where β ₀ is the maximum attractiveness factor, r _ij represents the distance between individual p _i and p _g , rand is a random number between 0 and 1, and the position x _i (t) of each individual in the group is expressed as ψ is the dihedral angle of amino acid residues in the input sequence, and L is the fragment length;

5.3) Perform L times of random fragment assembly for each individual in the group to complete the random swing of the group;

5.4) Recalculate the fluorescence brightness of each individual and update p _g ;

6) Determine whether the maximum number of iterations generation is reached;

6.1) If the current number of iterations is less than generation, return to step 5.1);

6.2) If the current number of iterations is equal to generation, end;

The technical idea of the present invention is: under the framework of the basic firefly algorithm, the coarse-grained energy model is used to effectively reduce the dimension of the conformational space, the population characteristics of the firefly algorithm are used to ensure the diversity of protein conformations, and the fragment assembly technology is used to initialize the conformational population , according to the coarse-grained expression model of protein conformation, the position of the conformation in space is represented by a set of dihedral angles, the energy ranking is used to determine the strongest luminescent individual, and the position of the conformation is updated by calculating the attractiveness between individuals, and finally in The near-native conformation with the minimum energy is searched in the conformational space.

The beneficial effects of the present invention are: the present invention is applied in protein structure prediction, and can obtain a conformation with high prediction accuracy and low complexity.

Description of drawings

Figure 1 is a three-dimensional schematic diagram of the closest conformation between the protein 2LOG predicted structure and the laboratory determined structure.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings.

With reference to Fig. 1, a kind of protein structure prediction based on firefly algorithm de novo method, described method comprises the following steps:

1) given input sequence information;

2) Parameter initialization: set the population size popSize, the number of iterations generation, the light intensity attraction factor γ, and the position update step factor α;

3) Population conformation initialization: According to the given input sequence, randomly generate popSize individuals, perform length fragment assembly for each individual in the population, and calculate its fluorescence brightness Io, where length is the sequence length, Io=-E, E is the protein conformational energy value calculated by the RosettaSscore3 energy function;

4) sort the fluorescence brightness calculated in step 3) from large to small, so that the individual with the largest fluorescence brightness is p _g ;

5) Start iteration:

5.1) For each individual in the group, calculate the attractiveness β of p _g to it;

5.2) Update the position of each individual in the space according to x _i (t+1)= _xi (t)+β(x _j (t) _–xi (t))+α(rand–0.5), where x _i (t+1), x _i (t) represents the updated position and current position of individual p _i , x _j (t) represents the current position of individual p _g , where β ₀ is the maximum attractiveness factor, r _ij represents the distance between individual p _i and p _g , rand is a random number between 0 and 1, and the position x _i (t) of each individual in the group is expressed as ψ is the dihedral angle of amino acid residues in the input sequence, and L is the fragment length;

5.3) Perform L times of random fragment assembly for each individual in the group to complete the random swing of the group;

5.4) Recalculate the fluorescence brightness of each individual and update p _g ;

6) Determine whether the maximum number of iterations generation is reached;

6.1) If the current number of iterations is less than generation, return to step 5.1);

6.2) If the current number of iterations is equal to generation, end;

This embodiment takes the test protein whose PDB name is 2LOG as an example, a method for protein structure prediction based on the firefly algorithm from scratch, and the method includes the following steps:

1) given input sequence information;

2) Parameter initialization: set the group size popSize=50, the number of iterations generation=10000, the light intensity attraction factor γ=0.5, and the location update step factor α=0.7;

3) Population conformation initialization: According to the given input sequence, randomly generate popSize individuals, perform length fragment assembly for each individual in the population, and calculate its fluorescence brightness Io, where length is the sequence length, Io=-E, E is the protein conformational energy value calculated by the RosettaSscore3 energy function;

4) sort the fluorescence brightness calculated in step 3) from large to small, so that the individual with the largest fluorescence brightness is p _g ;

5) Start iteration:

5.1) For each individual in the group, calculate the attractiveness β of p _g to it;

5.2) Update the position of each individual in the space according to x _i (t+1)= _xi (t)+β(x _j (t) _–xi (t))+α(rand–0.5), where x _i (t+1), x _i (t) represents the updated position and current position of individual p _i , x _j (t) represents the current position of individual p _g , where β ₀ is the maximum attractiveness factor, r _ij represents the distance between individual p _i and p _g , rand is a random number between 0 and 1, and the position x _i (t) of each individual in the group is expressed as ψ is the dihedral angle of amino acid residues in the input sequence, and L=3 is the fragment length;

5.3) Perform L times of random fragment assembly for each individual in the group to complete the random swing of the group;

5.4) Recalculate the fluorescence brightness of each individual and update p _g ;

6) Determine whether the maximum number of iterations generation is reached;

6.1) If the current number of iterations is less than generation, return to step 5.1);

6.2) If the current number of iterations is equal to generation, end;

What set forth above is the excellent effect shown by an embodiment of the present invention. Obviously, the present invention is not only suitable for the above-mentioned embodiment, but it can be used under the premise of not departing from the basic spirit of the present invention and not exceeding the content involved in the essence of the present invention. Make changes and implement them.

Claims (1)

1. a kind of protein structure prediction based on glowworm swarm algorithm from the beginning method it is characterised in that：Methods described includes following Step：

1) give list entries information；

2) parameter initialization：Setting population size popSize, iterationses generation, light intensity attracting factor γ, position is more New step factor α；

3) colony's conformation initialization：According to given list entries, random generate popSize individuality, in colony every each and every one Body does length fragment assembling, and calculates its fluorescent brightness Io, and wherein length is sequence length, and Io=-E, E are to pass through RosettaSscore3 energy function calculated protein conformation energy value；

4) to step 3) in the fluorescent brightness that calculates sort from big to small, make the maximum individuality of fluorescent brightness be p_g；

5) start iteration：

5.1) individual to each in colony, calculate p_gAttraction Degree β to it；

5.2) according to x_i(t+1)=x_i(t)+β(x_j(t)–x_i(t))+α (rand 0.5) updates each individuality position in space Put, wherein x_i(t+1), x_iT () represents individual p_iPosition after renewal and current position, x_jT () represents individual p_gPresent bit Put, whereinβ₀For the maximum Attraction Degree factor, r_ijRepresent individual p_iWith p_gThe distance between, rand be 0 to 1 it Between random number, each individual position x in colony_iT () is expressed as For the dihedral angle of the amino acid residue of list entries, L is fragment length；

5.3) each individuality in colony is carried out with L random fragment assembling, completes colony and swing at random；

5.4) recalculate each individual fluorescent brightness, update p_g；

6) judge whether to reach maximum iteration time generation；

6.1) if current iteration number of times is less than generation, return to step 5.1)；

6.2) if current iteration number of times is equal to generation, terminate.