Protein toxicity prediction method and system based on sequence information

Abstract

The invention provides a protein toxicity prediction method and system based on sequence information. The method comprises the following steps: acquiring a plurality of protein sequences from a protein database; performing feature calculation on all the protein sequences to obtain six types of protein feature vectors; splicing the six types of feature vectors according to columns to obtain a first feature vector; performing dimension reduction screening on the first feature vector to obtain a second feature vector of the target dimension; training a neural network model based on graph attention by using the second feature vector, and taking the trained neural network model as a protein toxicity prediction model; and inputting the new protein sequence into the protein toxicity prediction model to obtain a toxicity prediction result output by the protein toxicity prediction model. The method can adapt to different types of protein sequences, and the protein toxicity is accurately predicted through the learned universal sequence mode.

Description

Technical Field

[0001] The present invention belongs to the technical field of protein biology, and in particular relates to a protein toxicity prediction method and system based on sequence information. Background Art

[0002] Traditional methods of predicting protein toxicity mainly rely on biological experiments and structural analysis, and evaluate toxicity by studying the physicochemical properties or three-dimensional structure of proteins. However, these methods have many disadvantages. They often require a large amount of experimental data and resources. Not only are the experimental costs high, but the cycles are long. They are also highly dependent on experimental conditions, which makes it difficult to quickly and comprehensively predict the toxicity of a large number of unknown proteins. Based on this, how to use computational methods to efficiently and accurately predict toxicity has become a key research direction.

[0003] As biological sequence databases continue to expand, researchers have gradually shifted their focus from structural information to the direct use of sequence information. Sequence-based analysis methods can break through the limitations of obtaining protein structural data and use the information contained in the sequence itself to infer the biological function and toxicity of the protein. Initially, sequence-based prediction methods mainly used sequence alignment tools, such as BLAST, which was based on the principle of predicting toxicity by looking for sequence similarities with known toxic proteins. However, this sequence alignment method can only find similar sequences in known databases. Once faced with new sequences, variants, or proteins that have not been studied, its predictive ability will drop significantly.

[0004] With the development of deep learning technology, especially the successful application of convolutional neural networks (CNN) and recurrent neural networks (RNN), researchers have begun to explore how to use deep learning models to automatically extract potential features in protein sequences. CNN is widely used in the field of image processing and is good at capturing local feature patterns. When applied to protein sequences, it can effectively identify local patterns in amino acid sequences, such as the characteristics of the effects of specific sequence fragments or motifs on toxicity. RNN, especially its variants such as long short-term memory networks (LSTM) and gated recurrent units (GRU), has been applied to protein sequence toxicity prediction due to its good processing capabilities for sequence data. It can capture long-range dependencies in sequences, thereby better understanding the complex relationship between sequence information and toxicity.

[0005] Nevertheless, current methods using classical machine learning and traditional deep learning still have some shortcomings in protein toxicity prediction:

[0006] Feature extraction relies on manual design and cannot automatically capture complex features: In classic machine learning methods, feature extraction usually relies on manually designed features, such as amino acid composition, polarity, hydrophobicity, etc. These manually extracted features can often only reflect the surface information of protein sequences and are difficult to capture deeper sequence features, especially protein toxicity, which may be determined by small changes in the sequence, local structure, or long-range interactions. Due to the inability to automatically learn complex features from data, these methods show limitations when faced with potential patterns in sequences, resulting in insufficient prediction accuracy, and feature design often requires a lot of expertise and time, lacking flexibility and versatility.

[0007] The model's ability to model long-range dependencies in sequences is insufficient: Traditional deep learning methods, such as convolutional neural networks (CNNs), although they have certain advantages in capturing local features of protein sequences, are inadequate in dealing with long-range dependencies in sequences. The toxicity of proteins usually depends not only on the local region in the sequence, but is also closely related to the interaction between amino acid residues at a distance. Since CNN can only process a limited receptive field and cannot effectively model the global information of the entire sequence, it is difficult to fully capture the long-range dependency patterns in the sequence that affect toxicity, affecting the accurate prediction of complex proteins.

[0008] Insufficient generalization ability for unknown protein sequences and mutants: Classical machine learning and traditional deep learning methods usually rely on existing protein databases during training, and the generalization ability of the model is often limited to the scope of the training data. When faced with new sequences or mutated proteins that have not appeared in the training set, the prediction effect of the model is usually not ideal. This is because traditional models cannot fully learn the universal rules implicit in protein sequences, making it difficult for them to deal with new sequences or rare mutants outside the database, and unable to provide accurate toxicity predictions for these sequences. This lack of generalization ability limits the extensiveness of the model in actual biological research and applications. Summary of the invention

[0009] In view of the deficiencies in the prior art, the present invention provides a protein toxicity prediction method and system based on sequence information.

[0010] In a first aspect, the present invention provides a method for predicting protein toxicity based on sequence information, characterized by comprising:

[0011] Acquire multiple protein sequences from a protein database; wherein the multiple protein sequences include toxic and non-toxic protein sequences;

[0012] The features of all protein sequences were calculated to obtain the feature vectors of six categories of proteins;

[0013] The six categories of eigenvectors are concatenated column by column to obtain the first eigenvector;

[0014] Perform dimensionality reduction screening on the first eigenvector to obtain a second eigenvector of the target dimension;

[0015] The second eigenvector is used to train a graph attention-based neural network model, and the trained neural network model is used as a protein toxicity prediction model;

[0016] The new protein sequence is input into the protein toxicity prediction model to obtain the toxicity prediction result output by the protein toxicity prediction model.

[0017] Optionally, the feature calculation is performed on all protein sequences to obtain six categories of protein feature vectors, including:

[0018] Determine the feature vector of protein R;

[0019] Determining eigenvectors of a position-specific scoring matrix based on evolutionary features;

[0020] Determine feature vectors for amino acid indices based on physicochemical properties;

[0021] Determine the feature vector of the frequency counts of the amino acid group;

[0022] A feature vector that determines the length of the protein sequence;

[0023] Determine the feature vector of the common protein fragment sequence.

[0024] Optionally, determining the characteristic vector of protein R comprises:

[0025] The expression for constructing the characteristic vector AAC of amino acid composition is:

[0026] ;

[0027] Among them, p(a i ) is the ith amino acid a in the sequence i The frequency of occurrence of i ) is the ith amino acid a in the sequence i The number of occurrences; i=1,2,...,20; N is the total length of the sequence;

[0028] And / or, construct the expression of the dipeptide composition feature vector DPC:

[0029] ;

[0030] Among them, p(d i,j ) is the dipeptide d in the sequence i,j The frequency of occurrence of f(d i,j) is the dipeptide d in the sequence i,j The number of occurrences of; j=1,2,...,20;

[0031] And / or, construct an expression for the eigenvector MB of the Moreau-Broto autocorrelation:

[0032] ;

[0033] Among them, R d is the Moreau-Broto autocorrelation value of the sequence with delay d; d = 1, 2, ..., D; D represents the maximum delay; p i is the physicochemical property value of the ith amino acid in the sequence; p i+d is the physicochemical property value of the i+dth amino acid in the sequence;

[0034] And / or, construct the expression for the eigenvector MORAN of the Moran autocorrelation:

[0035] ;

[0036] Among them, M d is the Moran autocorrelation value of the sequence with a delay of d; is the mean of all physicochemical property values;

[0037] And / or, construct the expression for the eigenvector GEARY of the Geary autocorrelation:

[0038] ;

[0039] Among them, G d is the Geary autocorrelation value of the sequence with a delay of d;

[0040] And / or, construct the expression of the characteristic vector CTD of the combined transformation distribution:

[0041] ;

[0042] Among them, count(A i' ) is the i'th amino acid A i' The number of occurrences of Indicates the i'th amino acid A in the sequence i' and the j'th amino acid A j' The number of adjacencies; The representation captures the spatial distribution characteristics of each amino acid class; An operator representing the concatenation vector;

[0043] And / or, construct the expression for the eigenvector CTRIAD of the joint triples:

[0044] ;

[0045] in, Represents a triple The normalized frequency of occurrence of Represents a triple in a sequence The number of occurrences of g i is the group number of the ith amino acid in the sequence; g j is the group number of the jth amino acid in the sequence; is the group number of the k'th amino acid in the sequence;

[0046] And / or, construct the expression for the eigenvector SOCN of the sequence order coupling number:

[0047] ;

[0048] in, represents the number of sequential order couplings of delay d in the sequence on the physicochemical property k''; is the value of the physicochemical property k'' of the i-th amino acid; Representation and The amino acid property value of delay d; Represents the global variance of the attribute value, used for normalization; is the mean value of the physicochemical property k''; K is the total number of physicochemical properties;

[0049] And / or, construct the expression for the eigenvector QSO of the quasi-sequential order descriptor:

[0050] ;

[0051] in, It represents the coupling characteristics of the delay d in the sequence on the physicochemical property k'';

[0052] And / or, construct the expression of the feature vector PAAC of pseudo amino acid composition:

[0053] ;

[0054] in, represents the value of the physicochemical property k'' of the autocorrelation part of the sequence with delay d.

[0055] Optionally, determining a feature vector of a position-specific scoring matrix based on evolutionary features comprises:

[0056] The corrected probability of amino acid a at position i'' in the sequence is calculated according to the following formula :

[0057] ;

[0058] in, represents the frequency of occurrence of amino acid a at position i'' in the sequence; represents the frequency of occurrence of amino acid b at position i'' in the sequence; A is a set of natural amino acids;

[0059] The score of position i'' and amino acid a in the sequence is calculated according to the following formula :

[0060] ;

[0061] Where λ is a scaling factor used to adjust the scale of the score; q(a) is the average frequency of amino acid a in nature;

[0062] The average score p[a] of amino acid a was calculated according to the following formula:

[0063] ;

[0064] Where N is the total length of the sequence.

[0065] Optionally, the performing dimensionality reduction screening on the first feature vector to obtain a second feature vector of a target dimension includes:

[0066] Delete the columns with a standard deviation of 0 in the first eigenvector to obtain the third eigenvector;

[0067] Determine the Pearson correlation coefficient of each feature with the other features in the third feature vector;

[0068] Randomly delete one of the two features whose Pearson correlation coefficient is greater than the correlation coefficient threshold to obtain the fourth eigenvector;

[0069] constructing a co-selection relationship graph between features in the fourth feature vector to identify features associated with protein toxicity;

[0070] The features with the largest weights of the target number are selected as representatives and passed to recursive feature elimination based on cross-validation to obtain the second feature vector of the target dimension.

[0071] In a second aspect, the present invention provides a protein toxicity prediction system based on sequence information, comprising:

[0072] A sequence acquisition module is used to acquire multiple protein sequences from a protein database; wherein the multiple protein sequences include toxic and non-toxic protein sequences;

[0073] The feature calculation module is used to calculate the features of all protein sequences and obtain the feature vectors of six categories of proteins;

[0074] A vector concatenation module is used to concatenate the six types of feature vectors by column to obtain the first feature vector;

[0075] A dimension reduction screening module is used to perform dimension reduction screening on the first feature vector to obtain a second feature vector of a target dimension;

[0076] A model training module, used for training a graph attention-based neural network model using the second eigenvector, and using the trained neural network model as a protein toxicity prediction model;

[0077] The toxicity prediction module is used to input a new protein sequence into the protein toxicity prediction model to obtain a toxicity prediction result output by the protein toxicity prediction model.

[0078] Optionally, the feature calculation module includes:

[0079] A first determining unit, used to determine a characteristic vector of protein R;

[0080] A second determining unit is used to determine a feature vector of a position-specific scoring matrix based on evolutionary features;

[0081] A third determining unit is used to determine a feature vector of an amino acid index based on physicochemical properties;

[0082] A fourth determining unit, used to determine a characteristic vector of the frequency count of the amino acid group;

[0083] a fifth determining unit, for determining a feature vector of a length of a protein sequence;

[0084] The sixth determining unit is used to determine the feature vector of the protein common fragment sequence.

[0085] Optionally, the first determining unit includes:

[0086] The first construction means is used to construct the expression of the characteristic vector AAC of amino acid composition:

[0087] ;

[0088] Among them, p(a i ) is the ith amino acid a in the sequence i The frequency of occurrence of i ) is the ith amino acid a in the sequence i The number of occurrences; i=1,2,...,20; N is the total length of the sequence;

[0089] And / or, a second constructing means for constructing an expression of a characteristic vector DPC of a dipeptide composition:

[0090] ;

[0091] Among them, p(d i,j ) is the dipeptide d in the sequence i,j The frequency of occurrence of f(d i,j ) is the dipeptide d in the sequence i,j The number of occurrences of; j=1,2,...,20;

[0092] And / or, a third constructing means for constructing an expression for the characteristic vector MB of the Moreau-Broto autocorrelation:

[0093] ;

[0094] Among them, R d is the Moreau-Broto autocorrelation value of the sequence with delay d; d = 1, 2, ..., D; D represents the maximum delay; p i is the physicochemical property value of the ith amino acid in the sequence; p i+d is the physicochemical property value of the i+dth amino acid in the sequence;

[0095] And / or, a fourth constructing means for constructing an expression of the characteristic vector MORAN of Moran autocorrelation:

[0096] ;

[0097] Among them, M d is the Moran autocorrelation value of the sequence with a delay of d; is the mean of all physicochemical property values;

[0098] And / or, a fifth constructing means for constructing an expression for the characteristic vector GEARY of the Geary autocorrelation:

[0099] ;

[0100] Among them, G d is the Geary autocorrelation value of the sequence with a delay of d;

[0101] And / or, a sixth constructing means for constructing an expression of a characteristic vector CTD of a combined transformation distribution:

[0102] ;

[0103] Among them, count(A i' ) is the i'th amino acid A i' The number of occurrences of Indicates the i'th amino acid A in the sequence i' and the j'th amino acid A j'The number of adjacencies; The representation captures the spatial distribution characteristics of each amino acid class; An operator representing the concatenation vector;

[0104] And / or, a seventh constructing means for constructing an expression of a feature vector CTRIAD of a joint triple:

[0105] ;

[0106] in, Represents a triple The normalized frequency of occurrence of Represents a triple in a sequence The number of occurrences of g i is the group number of the ith amino acid in the sequence; g j is the group number of the jth amino acid in the sequence; is the group number of the k'th amino acid in the sequence;

[0107] And / or, an eighth constructing means for constructing an expression of a characteristic vector SOCN of a sequence order coupling number:

[0108] ;

[0109] in, represents the number of sequential order couplings of delay d in the sequence on the physicochemical property k''; is the value of the physicochemical property k'' of the i-th amino acid; Representation and The amino acid property value of delay d; Represents the global variance of the attribute value, used for normalization; is the mean value of the physicochemical property k''; K is the total number of physicochemical properties;

[0110] And / or, a ninth constructing means for constructing an expression of a characteristic vector QSO of a quasi-sequence order descriptor:

[0111] ;

[0112] in, It represents the coupling characteristics of the delay d in the sequence on the physicochemical property k'';

[0113] And / or, a tenth construction means for constructing an expression of a characteristic vector PAAC of pseudo amino acid composition:

[0114] ;

[0115] in, represents the value of the physicochemical property k'' of the autocorrelation part of the sequence with delay d.

[0116] Optionally, the second determining unit includes:

[0117] The first calculation device is used to calculate the correction probability of amino acid a at position i'' in the sequence according to the following formula :

[0118] ;

[0119] in, represents the frequency of occurrence of amino acid a at position i'' in the sequence; represents the frequency of occurrence of amino acid b at position i'' in the sequence; A is a set of natural amino acids;

[0120] The second calculation device is used to calculate the score of position i'' and amino acid a in the sequence according to the following formula :

[0121] ;

[0122] Where λ is a scaling factor used to adjust the scale of the score; q(a) is the average frequency of amino acid a in nature;

[0123] The third calculating means is used to calculate the average score p[a] of amino acid a according to the following formula:

[0124] ;

[0125] Where N is the total length of the sequence.

[0126] Optionally, the dimension reduction screening module includes:

[0127] A first deletion unit is used to delete the columns with a standard deviation of 0 in the first eigenvector to obtain a third eigenvector;

[0128] A determination unit, used to determine the Pearson correlation coefficient between each feature in the third feature vector and other features;

[0129] The second deletion unit is used to randomly delete one of the two features whose Pearson correlation coefficient is greater than the correlation coefficient threshold, so as to obtain a fourth feature vector;

[0130] a construction unit for constructing a co-selection relationship graph between features in the fourth feature vector to identify features associated with protein toxicity;

[0131] The selection unit is used to select the target number of features with the largest weight as representatives, and pass them to the recursive feature elimination based on cross-validation to obtain the second feature vector of the target dimension.

[0132] The present invention provides a method and system for predicting protein toxicity based on sequence information. The method introduces a deep learning model (such as a neural network model based on a self-attention mechanism) that can extract multi-level and deep features from the original sequence, especially capture features that are difficult to design or cannot be discovered manually, improve the prediction accuracy of the model, and reduce dependence on artificial feature design. It can adapt to different types of protein sequences.

[0133] The method combines a pre-trained language model with protein sequence embedding technology. By pre-training on large-scale unlabeled protein sequence data, rich sequence representations are learned, thereby providing a powerful basic model for subsequent toxicity prediction. The model's predictive ability on new protein sequences is improved. Even for variants or rare proteins that have never been seen before, the method of the present invention can make accurate predictions through the learned general sequence patterns. BRIEF DESCRIPTION OF THE DRAWINGS

[0134] In order to more clearly illustrate the technical solution of the present invention, the drawings required for use in the embodiments are briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0135] Figure 1 A schematic diagram of a flow chart of a method for predicting protein toxicity based on sequence information provided in an embodiment of the present invention;

[0136] Figure 2 A schematic diagram of the structure of a deep learning model provided by an embodiment of the present invention;

[0137] Figure 3 A comparison chart of the classification performance of the embodiments of the present invention with other classifiers on a test set;

[0138] Figure 4 A schematic diagram of the structure of a protein toxicity prediction system based on sequence information provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0139] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention. Example 1

[0140] like Figure 1As shown, this embodiment provides a method for predicting protein toxicity based on sequence information, comprising:

[0141] Step 101, obtaining a plurality of protein sequences from a protein database; wherein the plurality of protein sequences include toxic and non-toxic protein sequences.

[0142] In this step, the keyword "KW-0800 AND Reviewed" is used to search from the complete database on the protein database website. The keyword "KW-0800" represents naturally occurring toxic proteins that damage or kill other cells. These toxic proteins are produced by some poisonous animals (mainly from snakes, scorpions, spiders, anemones and cone snails), plants, fungi and pathogenic bacteria. "Reviewed" represents data that has been manually reviewed and experimentally confirmed. The protein sequences obtained using the "KW-0800AND Reviewed" keyword are used as a set of toxic positive samples (also called positive samples).

[0143] Then, the non-toxic, non-allergenic and manually reviewed protein sequences were retrieved using the keywords “NOT KW-0800 NOT KW-0020 AND Reviewed”. The resulting protein sequences were used as a set of non-toxic negative samples (also called negative samples).

[0144] The two groups of protein sequence samples are stored in two FASTA files. FASTA is a text file format widely used in bioinformatics, mainly used to store nucleic acid sequences (such as DNA, RNA) or protein sequences and their related annotation information.

[0145] The obtained protein sequences need to be strictly screened to ensure the reliability and accuracy of subsequent analysis. First, the lengths of all sequences of positive and negative samples, i.e., toxic and non-toxic sample sets, are checked, and sequences shorter than 35 amino acids are removed. Too short sequences are usually fragmented or incomplete peptide sequences (short-chain proteins are usually called peptides), which cannot provide enough information to support analysis tasks such as function prediction or sequence alignment, so too short protein sequences are excluded.

[0146] This embodiment also filters the ambiguous amino acids in the sequence. Ambiguous amino acids are usually represented by the symbols B (indicating the unknown state of aspartic acid or asparagine), J (indicating the unknown state of leucine or isoleucine), O (indicating pyrrolysine, a rare amino acid, often introduced into proteins as a result of post-translational modification), U (indicating selenocysteine, introduced by a specific signal, which is very rare), X (unknown amino acid), and Z (indicating the ambiguity of glutamate or glutamine). These ambiguous symbols may interfere with the accuracy of subsequent analysis, so all protein sequences containing these symbols are removed from the data set. After this step of data cleaning, the retained protein sequences meet the length requirements and do not contain any ambiguous amino acids.

[0147] Step 102, perform feature calculation on all protein sequences to obtain feature vectors of six major categories of proteins.

[0148] Exemplarily, this step includes determining a feature vector of protein R.

[0149] Determine eigenvectors of a position-specific scoring matrix based on evolutionary signatures.

[0150] Determine feature vectors for amino acid indices based on physicochemical properties.

[0151] Determine the feature vector of frequency counts for the amino acid groups.

[0152] A feature vector that determines the length of a protein sequence.

[0153] Determine the feature vector of the common protein fragment sequence.

[0154] Among them, the characteristic vector for determining protein R includes:

[0155] The expression for constructing the characteristic vector AAC of amino acid composition is:

[0156] .

[0157] Among them, p(a i ) is the ith amino acid a in the sequence i The frequency of occurrence of i ) is the ith amino acid a in the sequence i The number of occurrences; i=1,2,...,20; N is the total length of the sequence; in this embodiment, only 20 common amino acids in nature are considered, so the length of the vector AAC is fixed to 20, and each value represents the relative frequency of the corresponding amino acid.

[0158] And / or, construct the expression of the dipeptide composition feature vector DPC:

[0159] .

[0160] Among them, p(d i,j ) is the dipeptide d in the sequence i,j The frequency of occurrence of f(d i,j ) is the dipeptide d in the sequence i,j The number of occurrences of; j = 1, 2, ..., 20; a sequence of length N has N-1 adjacent amino acid pairs, so N-1 represents the total number of dipeptides that may be formed in the sequence; the length of the vector is fixed to 400, and each value represents the relative frequency of the corresponding dipeptide.

[0161] And / or, construct an expression for the eigenvector MB of the Moreau-Broto autocorrelation:

[0162] .

[0163] Among them, R d is the Moreau-Broto autocorrelation value of the sequence with a delay of d; d represents the number of amino acids in the autocorrelation calculation, which is called the delay; d=1,2,...,D; D represents the maximum delay, and in this embodiment, D takes a fixed value of 240; p i is the physicochemical property value of the ith amino acid in the sequence; p i+d is the physicochemical property value of the i+dth amino acid in the sequence.

[0164] And / or, construct the expression for the eigenvector MORAN of the Moran autocorrelation:

[0165] .

[0166] Among them, M d is the Moran autocorrelation value of the sequence with a delay of d; is the mean of all physicochemical property values;

[0167] And / or, construct the expression for the eigenvector GEARY of the Geary autocorrelation:

[0168] .

[0169] Among them, G d is the Geary autocorrelation value of the sequence with a delay of d;

[0170] And / or, construct the expression of the characteristic vector CTD of the combined transformation distribution:

[0171] .

[0172] Among them, count(A i' ) is the i'th amino acid A i' The number of occurrences of Indicates the i'th amino acid A in the sequence i' and the j'th amino acid A j' In this embodiment, considering the rare 21st amino acid, the values ​​of i' and j' range from 1 to 21; The representation captures the spatial distribution characteristics of each amino acid class; Operator representing the splicing vector; position encoding is performed at specific positions in the sequence (the first occurrence position, the 25th position, the 50th position, the 75th position, and the last position), and 5 distribution positions are calculated for each category. The 21st rare amino acid is also considered here, so the feature vector dimension of CTD is 21+21+105=147.

[0173] And / or, construct the expression for the eigenvector CTRIAD of the joint triples:

[0174] .

[0175] The physical and chemical properties of amino acids are grouped and mapped as G={g1,g2,…,g N}; g i =1,2,…,7; g i Indicates the group number of the i-th amino acid; each component of CTRIAD represents a specific triple The normalized frequency of ; Represents a triple in a sequence The number of occurrences of g i is the group number of the ith amino acid in the sequence; g j is the group number of the jth amino acid in the sequence; is the group number of the k'th amino acid in the sequence.

[0176] And / or, construct the expression for the eigenvector SOCN of the sequence order coupling number:

[0177] .

[0178] in, represents the number of sequential order couplings of delay d in the sequence on the physicochemical property k''; is the value of the physicochemical property k'' of the i-th amino acid; Representation and The amino acid property value of delay d; Represents the global variance of the attribute value, used for normalization; is the mean of the physical and chemical properties k''; K is the total number of physical and chemical properties; in this embodiment, the number of physical and chemical properties considered K is 2; the maximum delay considered D is 30, therefore, the length of the SOCN feature vector is K×D=60.

[0179] And / or, construct the expression for the eigenvector QSO of the quasi-sequential order descriptor:

[0180] .

[0181] in, Represents the coupling characteristics of the delay d in the sequence on the physical and chemical property k''; in the embodiment, QSO is a complete feature vector with a total length of 100.

[0182] And / or, construct the expression of the feature vector PAAC of pseudo amino acid composition:

[0183] .

[0184] in, represents the value of the physicochemical property k'' of the autocorrelation part of the sequence with delay d.

[0185] In this embodiment, the feature vector of pseudo amino acid composition (K=3, D=10) only considers limited delayed autocorrelation information of primary sequence characteristics and physicochemical properties. The feature vector APAAC (i.e. K=6, D=10 in PAAC) of amphipathic pseudo amino acid composition is extended to include richer physicochemical properties and higher-level autocorrelation information.

[0186] Determine the eigenvectors of the position-specific scoring matrix based on the evolutionary features, including:

[0187] Use the tool Psi-BLAST to calculate the position-specific scoring matrix (hereinafter referred to as PSSM) for each protein sample. Set the search database to the database file after uniprot review, the threshold evalue value is 0.001, and the number of iterations is 3. First, Psi-BLAST starts with a query sequence and performs a standard BLAST search to find sequences that are significantly similar to the query sequence. Secondly, align the searched homologous sequences (usually filtered by the evalue threshold) with the query sequence to generate a multiple sequence alignment. Next, calculate the position-specific frequency matrix.

[0188] The corrected probability of amino acid a at position i'' in the sequence is calculated according to the following formula :

[0189] .

[0190] in, represents the frequency of occurrence of amino acid a at position i'' in the sequence; represents the frequency of occurrence of amino acid b at position i'' in the sequence; A is a set of natural amino acids (20 natural amino acids).

[0191] The score of position i'' and amino acid a in the sequence is calculated according to the following formula , which are the elements of PSSM:

[0192] .

[0193] Among them, λ is a scaling factor used to adjust the scale of the score; q(a) is the average frequency of amino acid a in nature, that is, the background probability, which is a value based on the statistics of the reference database.

[0194] Psi-BLAST uses the updated PSSM to re-search the database to find new homologous sequences, repeating the above process until the results converge, and finally, using the global average pooling algorithm.

[0195] The average score p[a] of amino acid a was calculated according to the following formula:

[0196] .

[0197] Where N is the total length of the sequence; in this embodiment, P[a] is the average score of amino acid a in the global representation after pooling. This operation is independent of input length and is very suitable for unified processing of variable-length inputs. The feature vector length is 20 dimensions.

[0198] Amino acid index based on physicochemical properties. The physicochemical properties of 617 amino acids, including amino acid volume, folding tendency, isoelectric point, flexibility, and folding free energy, were selected from the database AAindex. The physicochemical properties of each protein sequence were statistically analyzed, and the length of the feature vector dimension was 617 dimensions.

[0199] Amino acid group frequency count. Amino acids are divided into 6 categories: hydrophobic, positively charged, negatively charged, polar, conformational, and others. The amino acids in each sequence are grouped and the number of amino acids in each group is counted. The length of the feature vector dimension obtained is 6 dimensions.

[0200] Protein sequence length statistics: Just count the number of amino acids, and the feature vector length is 1 dimension.

[0201] Common fragment sequence search. Input the negative set samples and the positive set samples into the MERCI tool respectively, mine possible common fragment sequences from the positive set samples, and use the negative set samples to verify the specificity of the common fragment sequences. The core goal of MERCI is to find common fragment sequences that are significantly enriched in the positive set samples and almost non-existent in the negative set samples. Set the number of common fragment sequences to 50, and use the one-hot encoding method to generate a 50-dimensional feature vector from the common fragment sequences.

[0202] Step 103, concatenate the six types of feature vectors by column to obtain the first feature vector.

[0203] All feature vectors obtained in step 102 are concatenated together according to dimension 1 to obtain the first feature vector Features:

[0204] .

[0205] Step 104: perform dimensionality reduction screening on the first eigenvector to obtain a second eigenvector of the target dimension.

[0206] Exemplarily, this step includes:

[0207] Delete the columns with standard deviation of 0 in the first eigenvector to obtain the third eigenvector.

[0208] Determine the Pearson correlation coefficient of each feature with the other features in the third eigenvector.

[0209] If the correlation coefficient of two features is greater than or equal to 0.8, the two features are strongly correlated. Therefore, one of the two features whose Pearson correlation coefficient is greater than the correlation coefficient threshold (0.8 in this embodiment) is randomly deleted to obtain the fourth feature vector.

[0210] A co-selection relationship graph between features in the fourth feature vector was constructed to identify features associated with protein toxicity.

[0211] The features with the largest weights of the target number are selected as representatives and passed to recursive feature elimination based on cross-validation to obtain the second feature vector of the target dimension.

[0212] By running a variety of machine learning algorithms (such as decision trees, support vector machines, gradient boosting machines, etc.), we build a co-selection relationship graph between features and identify key features that are highly correlated with protein toxicity. We set the 100 features with the largest weights as representatives and pass them to recursive feature elimination based on cross-validation. The goal is to find an optimal feature subset for a given feature matrix and target to optimize model performance. The evaluation function is:

[0213] .

[0214] Among them, K f is the number of cross-validation folds, which is selected as 20 folds; is the feature vector of the hth feature to be trained; is the feature vector of the hth feature to be verified; y (train) is the training label; y (val) The inference result label for verification. Recursively eliminate features until the optimal one is found.

[0215] Step 105, using the second eigenvector to train a graph attention-based neural network model, and using the trained neural network model as a protein toxicity prediction model.

[0216] A dual-channel graph attention network is used to extract the protein feature matrix. The model architecture is as follows Figure 2 As shown in Figure 1. The feature data is first preprocessed and subjected to one-dimensional convolution operations to generate a preliminary feature representation. The one-dimensional convolution layer can effectively capture the relationship between local adjacent amino acids in the sequence, which helps to extract local patterns in the protein (such as toxic fragments). The output of this convolution layer will be used in the subsequent feature extraction module. The model uses two parallel dual-channel graph attention network modules. One is a feature-oriented attention module, which receives the feature representation, establishes associations between feature nodes through the graph attention mechanism, and outputs feature-oriented embeddings. This module focuses on the interaction of protein features themselves and identifies important toxicity-related fragments in the sequence, such as specific amino acid combinations or structures; the other is a sequence-oriented graph attention module, which captures long-range dependencies at different positions in the sequence by constructing a sequence-oriented graph structure and outputs sequence-oriented embeddings. This mechanism helps the model understand the complex interactions in the protein sequence, especially the relationships between long-distance amino acids, and helps identify complex toxic fragments. After feature fusion, the concatenated feature vector is further processed by the gated recurrent unit GRU module to capture the complex correlations in the protein sequence. GRU can extract deeper dependencies and patterns in the sequence, helping the model to better identify toxicity-related feature combinations. The output vector of GRU will be used as the input of the subsequent classification module. The classifier module generates an inference score between 0 and 1 for the final toxicity prediction decision. If the inference score is greater than or equal to a certain threshold (set to 0.5), the protein is judged to be toxic, otherwise it is non-toxic. The model training step will save a trained model as a protein toxicity prediction model for subsequent inference prediction of new samples.

[0217] Step 106: input the new protein sequence into the protein toxicity prediction model to obtain a toxicity prediction result output by the protein toxicity prediction model.

[0218] Input a new protein sequence file in FASTA format to the protein toxicity prediction model and directly perform inference prediction. The model gives a corresponding comma-delimited file, annotating the toxicity of the input protein.

[0219] In order to verify the prediction performance of the toxicity prediction method provided in this embodiment, the test was conducted on the uniprotKB dataset and compared with other prediction tools, including ToxinPred2, ToxIBTL, Toxify, CSM-Toxin, VISH-Pred and ToxinPred3. The performance comparison results are shown in Figure 2. Figure 3 shown.

[0220] Figure 3 In the figure, the bold indicates the best, and the underline indicates the second best. According to the experimental results, our invented method is superior to other existing classification predictors in multiple indicators such as accuracy, recall rate, F1 value, area under the ROC curve AUC value, Matthews correlation coefficient, etc. These performance advantages fully demonstrate that this method has significant classification effect and reliability in protein toxicity prediction tasks.

[0221] The accuracy of the toxicity prediction method provided in this embodiment reaches 0.948, which is significantly higher than other predictors, especially about 5.5% higher than the current optimal VISH-Pred, showing its significant advantage in overall classification accuracy. In terms of recall rate, the recall rate of the toxicity prediction method provided in this embodiment reaches 0.953, which is significantly improved compared with other methods, indicating that the toxicity prediction method provided in this embodiment can more effectively capture positive samples in protein toxicity and improve the sensitivity of the model. In terms of the comprehensive indicator of F1 value, the F1 value of the toxicity prediction method provided in this embodiment is 0.932, which is much higher than other models, indicating that the model can accurately predict toxic proteins while also reducing the occurrence of false positives and maintaining good balanced performance. In terms of AUC, the AUC value of the toxicity prediction method provided in this embodiment reaches 0.949, which is nearly 7.8% higher than VISH-Pred's 0.871, showing a stronger ability to distinguish positive and negative samples at different thresholds, enhancing the robustness of the model. In terms of MCC, the MCC value of the toxicity prediction method provided in this embodiment reached 0.891, which further proved that the model still has excellent prediction performance when the sample categories are unbalanced.

[0222] In summary, this embodiment provides a protein toxicity prediction method based on sequence information, which automatically extracts features from protein sequences by introducing deep learning models (such as convolutional neural networks, recurrent neural networks, and Transformer models based on self-attention mechanisms). Unlike traditional methods, this embodiment utilizes the powerful feature learning capabilities of deep learning models to extract multi-level and deep features from the original sequence, especially to capture features that are difficult to design or cannot be discovered manually. This automation of feature extraction greatly improves the prediction accuracy of the model, while reducing dependence on manual feature design, making this embodiment more versatile and adaptable to different types of protein sequences.

[0223] Modeling global and local dependencies of sequences: Existing deep learning models, such as convolutional neural networks (CNNs), although they have certain advantages in capturing local features of protein sequences, have difficulty in handling long-range dependencies in sequences. The toxicity of proteins is often not only related to local sequence fragments, but is also affected by complex interactions between long-distance amino acid residues. In order to overcome this limitation, this embodiment adopts more advanced deep learning architectures, such as Transformer or improved RNN models, which can process the entire sequence through the self-attention mechanism (Self-Attention), thereby effectively modeling long-range dependencies in protein sequences. At the same time, this embodiment introduces a multi-scale feature fusion mechanism in the model, which can capture local and global feature information at the same time, ensuring that toxicity prediction not only takes into account local patterns in the sequence, but also integrates the interactions between long-distance amino acids, thereby improving the comprehensiveness and accuracy of the prediction.

[0224] Enhance the generalization ability of unknown protein sequences and variants: The existing technology often lacks sufficient generalization ability for the prediction of new protein sequences or variants. The model is too dependent on known databases during training, and the amount of data in these databases is limited and the coverage is not comprehensive. When faced with sequences that have not appeared in the training set, the prediction performance of traditional models is significantly reduced. In order to solve this problem, this embodiment combines pre-trained language models with protein sequence embedding technology. These pre-trained models learn rich sequence representations by pre-training on large-scale unlabeled protein sequence data, thereby providing a powerful basic model for subsequent toxicity prediction. This method greatly improves the prediction ability of the model on new protein sequences. Even for variants or rare proteins that have never been seen before, the method of this embodiment can accurately predict through the learned general sequence patterns. In addition, this embodiment also uses data enhancement techniques, such as random sequence perturbations and mutation simulations, to further enhance the adaptability of the model when processing variant sequences. Example 2

[0225] Based on the same inventive concept as Example 1, this example provides a protein toxicity prediction system based on sequence information. Since the principle of solving the problem by this system is similar to the aforementioned protein toxicity prediction method based on sequence information, the implementation of this system can refer to the implementation of the protein toxicity prediction method based on sequence information.

[0226] like Figure 4 As shown, the protein toxicity prediction system based on sequence information includes:

[0227] The sequence acquisition module 10 is used to acquire multiple protein sequences from a protein database; wherein the multiple protein sequences include toxic and non-toxic protein sequences.

[0228] The feature calculation module 20 is used to perform feature calculation on all protein sequences to obtain feature vectors of six major categories of proteins.

[0229] The vector concatenation module 30 is used to concatenate the six types of feature vectors by column to obtain a first feature vector.

[0230] The dimension reduction and screening module 40 is used to perform dimension reduction and screening on the first feature vector to obtain a second feature vector of a target dimension.

[0231] The model training module 50 is used to train a graph attention-based neural network model using the second eigenvector, and use the trained neural network model as a protein toxicity prediction model.

[0232] The toxicity prediction module 60 is used to input a new protein sequence into the protein toxicity prediction model to obtain a toxicity prediction result output by the protein toxicity prediction model.

[0233] Exemplarily, the feature calculation module includes:

[0234] The first determining unit is used to determine the characteristic vector of protein R.

[0235] The second determining unit is used to determine a feature vector of a position-specific scoring matrix based on evolutionary features.

[0236] The third determining unit is used to determine a feature vector of the amino acid index based on the physicochemical properties.

[0237] The fourth determining unit is used to determine a feature vector of the amino acid group frequency count.

[0238] The fifth determining unit is used to determine the feature vector of the protein sequence length.

[0239] The sixth determining unit is used to determine the feature vector of the protein common fragment sequence.

[0240] Exemplarily, the first determining unit includes:

[0241] The first construction means is used to construct the expression of the characteristic vector AAC of amino acid composition:

[0242] .

[0243] Among them, p(a i ) is the ith amino acid a in the sequence i The frequency of occurrence of i ) is the ith amino acid a in the sequence i The number of occurrences of; i=1,2,...,20; N is the total length of the sequence.

[0244] And / or, a second constructing means for constructing an expression of a characteristic vector DPC of a dipeptide composition:

[0245] .

[0246] Among them, p(d i,j ) is the dipeptide d in the sequence i,j The frequency of occurrence of f(d i,j ) is the dipeptide d in the sequence i,j The number of occurrences of ; j=1,2,...,20.

[0247] And / or, a third constructing means for constructing an expression for the characteristic vector MB of the Moreau-Broto autocorrelation:

[0248] .

[0249] Among them, R d is the Moreau-Broto autocorrelation value of the sequence with delay d; d = 1, 2, ..., D; D represents the maximum delay; p i is the physicochemical property value of the ith amino acid in the sequence; p i+d is the physicochemical property value of the i+dth amino acid in the sequence.

[0250] And / or, a fourth constructing means for constructing an expression of the characteristic vector MORAN of Moran autocorrelation:

[0251] .

[0252] Among them, M d is the Moran autocorrelation value of the sequence with a delay of d; is the mean of all physicochemical property values.

[0253] And / or, a fifth constructing means for constructing an expression for the characteristic vector GEARY of the Geary autocorrelation:

[0254] .

[0255] Among them, G d is the Geary autocorrelation value of the sequence with a delay of d.

[0256] And / or, a sixth constructing means for constructing an expression of a characteristic vector CTD of a combined transformation distribution:

[0257] .

[0258] Among them, count(A i' ) is the i'th amino acid A i' The number of occurrences of Indicates the i'th amino acid A in the sequence i' and the j'th amino acid A j' The number of adjacencies; The representation captures the spatial distribution characteristics of each amino acid class; An operator that represents concatenation of vectors.

[0259] And / or, a seventh constructing means for constructing an expression of a feature vector CTRIAD of a joint triple:

[0260] .

[0261] in, Represents a triple The normalized frequency of occurrence of Represents a triple in a sequence The number of occurrences of g i is the group number of the ith amino acid in the sequence; g j is the group number of the jth amino acid in the sequence; is the group number of the k'th amino acid in the sequence.

[0262] And / or, an eighth constructing means for constructing an expression of a characteristic vector SOCN of a sequence order coupling number:

[0263] .

[0264] in, represents the number of sequential order couplings of delay d in the sequence on the physicochemical property k''; is the value of the physicochemical property k'' of the i-th amino acid; Representation and The amino acid property value of delay d; Represents the global variance of the attribute value, used for normalization; is the mean of the physicochemical properties k''; K is the total number of physicochemical properties.

[0265] And / or, a ninth constructing means for constructing an expression of a characteristic vector QSO of a quasi-sequence order descriptor:

[0266] .

[0267] in, It represents the coupling characteristics of the delay d in the sequence on the physicochemical property k'';

[0268] And / or, a tenth construction means for constructing an expression of a characteristic vector PAAC of pseudo amino acid composition:

[0269] .

[0270] in, represents the value of the physicochemical property k'' of the autocorrelation part of the sequence with delay d.

[0271] Exemplarily, the second determining unit includes:

[0272] The first calculation device is used to calculate the correction probability of amino acid a at position i'' in the sequence according to the following formula :

[0273] .

[0274] in, represents the frequency of occurrence of amino acid a at position i'' in the sequence; represents the frequency of occurrence of amino acid b at position i'' in the sequence; A is a set of natural amino acids.

[0275] The second calculation device is used to calculate the score of position i'' and amino acid a in the sequence according to the following formula :

[0276] .

[0277] Among them, λ is the scaling factor used to adjust the scale of the score; q(a) is the average frequency of amino acid a in nature.

[0278] The third calculating means is used to calculate the average score p[a] of amino acid a according to the following formula:

[0279] .

[0280] Where N is the total length of the sequence.

[0281] Exemplarily, the dimension reduction screening module includes:

[0282] The first deletion unit is used to delete the columns with a standard deviation of 0 in the first eigenvector to obtain a third eigenvector.

[0283] The determination unit is used to determine the Pearson correlation coefficient between each feature in the third feature vector and other features.

[0284] The second deletion unit is used to randomly delete one of the two features whose Pearson correlation coefficient is greater than the correlation coefficient threshold, so as to obtain a fourth feature vector;

[0285] A construction unit is used to construct a co-selection relationship map between features in the fourth feature vector to identify features associated with protein toxicity.

[0286] The selection unit is used to select the target number of features with the largest weight as representatives, and pass them to the recursive feature elimination based on cross-validation to obtain the second feature vector of the target dimension.

[0287] For more specific working processes of the above modules, please refer to the corresponding contents disclosed in Example 1, which will not be repeated here. Example 3

[0288] This embodiment provides a computer device, including a processor and a memory; wherein the processor implements the steps of the protein toxicity prediction method based on sequence information described in Example 1 when executing a computer program stored in the memory.

[0289] For more specific details of the above method, please refer to the corresponding contents disclosed in Example 1, which will not be repeated here. Example 4

[0290] This embodiment provides a computer-readable storage medium for storing a computer program; when the computer program is executed by a processor, the steps of the protein toxicity prediction method based on sequence information described in Embodiment 1 are implemented.

[0291] For more specific details of the above method, please refer to the corresponding contents disclosed in Example 1, which will not be repeated here. Example 5

[0292] This embodiment provides a computer program product, including computer executable instructions or a computer program. When the computer executable instructions or the computer program are executed by a processor, the steps of the protein toxicity prediction method based on sequence information described in Example 1 are implemented.

[0293] For more specific details of the above method, please refer to the corresponding contents disclosed in Example 1, which will not be repeated here.

[0294] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the systems, devices, storage media, and computer program products disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0295] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on this understanding, the technical solution in the embodiments of the present invention can be essentially or partly contributed to the prior art in the form of a software product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present invention or certain parts of the embodiments.

[0296] In some embodiments, computer executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.

[0297] As an example, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file storing other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).

[0298] As an example, computer executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed at multiple sites and interconnected by a communication network.

[0299] The present invention has been described in detail above in conjunction with specific implementations and exemplary examples, but these descriptions cannot be understood as limiting the present invention. Those skilled in the art understand that, without departing from the spirit and scope of the present invention, a variety of equivalent substitutions, modifications or improvements may be made to the technical solution of the present invention and its implementation methods, all of which fall within the scope of the present invention. The scope of protection of the present invention shall be subject to the attached claims.

Claims

1. A method for predicting protein toxicity based on sequence information, characterized in that: include: Acquire multiple protein sequences from a protein database; wherein the multiple protein sequences include toxic and non-toxic protein sequences; The features of all protein sequences were calculated to obtain the feature vectors of six categories of proteins; The six categories of eigenvectors are concatenated column by column to obtain the first eigenvector; Perform dimensionality reduction screening on the first eigenvector to obtain a second eigenvector of the target dimension; The second eigenvector is used to train a graph attention-based neural network model, and the trained neural network model is used as a protein toxicity prediction model; The new protein sequence is input into the protein toxicity prediction model to obtain the toxicity prediction result output by the protein toxicity prediction model.

2. The protein toxicity prediction method according to claim 1, characterized in that: The feature calculations for all protein sequences are performed to obtain six categories of protein feature vectors, including: Determine the feature vector of protein R; Determining eigenvectors of a position-specific scoring matrix based on evolutionary features; Determine feature vectors for amino acid indices based on physicochemical properties; Determine the feature vector of the frequency counts of the amino acid group; A feature vector that determines the length of the protein sequence; Determine the feature vector of the common protein fragment sequence.

3. The protein toxicity prediction method according to claim 2, characterized in that: The step of determining the characteristic vector of protein R comprises: The expression for constructing the characteristic vector AAC of amino acid composition is: ； Among them, p(a i ) is the ith amino acid a in the sequence i The frequency of occurrence of i ) is the ith amino acid a in the sequence i The number of occurrences; i=1,2,...,20; N is the total length of the sequence; And / or, construct the expression of the dipeptide composition feature vector DPC: ； Among them, p(d i,j ) is the dipeptide d in the sequence i,j The frequency of occurrence of f(d i,j ) is the dipeptide d in the sequence i,j The number of occurrences of; j=1,2,...,20; And / or, construct an expression for the eigenvector MB of the Moreau-Broto autocorrelation: ； Among them, R d is the Moreau-Broto autocorrelation value of the sequence with delay d; d = 1, 2, ..., D; D represents the maximum delay; p i is the physicochemical property value of the ith amino acid in the sequence; p i+d is the physicochemical property value of the i+dth amino acid in the sequence; And / or, construct the expression for the eigenvector MORAN of the Moran autocorrelation: ； Among them, M d is the Moran autocorrelation value of the sequence with a delay of d; is the mean of all physicochemical property values; And / or, construct the expression for the eigenvector GEARY of the Geary autocorrelation: ； Among them, G d is the Geary autocorrelation value of the sequence with a delay of d; And / or, construct the expression of the characteristic vector CTD of the combined transformation distribution: ； Among them, count(A i' ) is the i'th amino acid A i' The number of occurrences of Indicates the i'th amino acid A in the sequence i' and the j'th amino acid A j' The number of adjacencies; The representation captures the spatial distribution characteristics of each amino acid class; An operator representing the concatenation vector; And / or, construct the expression for the eigenvector CTRIAD of the joint triples: ； in, Represents a triple The normalized frequency of occurrence of Represents a triple in a sequence The number of occurrences of g i is the group number of the ith amino acid in the sequence; g j is the group number of the jth amino acid in the sequence; is the group number of the k'th amino acid in the sequence; And / or, construct the expression for the eigenvector SOCN of the sequence order coupling number: ； in, represents the number of sequential order couplings of delay d in the sequence on the physicochemical property k''; is the value of the physicochemical property k'' of the i-th amino acid; Representation and The amino acid property value of delay d; Represents the global variance of the attribute value, used for normalization; is the mean value of the physicochemical property k''; K is the total number of physicochemical properties; And / or, construct the expression for the eigenvector QSO of the quasi-sequential order descriptor: ； in, It represents the coupling characteristics of the delay d in the sequence on the physicochemical property k''; And / or, construct the expression of the feature vector PAAC of pseudo amino acid composition: ； in, represents the value of the physicochemical property k'' of the autocorrelation part of the sequence with delay d.

4. The protein toxicity prediction method according to claim 2, characterized in that: The step of determining a feature vector of a position-specific scoring matrix based on evolutionary features comprises: The corrected probability of amino acid a at position i'' in the sequence is calculated according to the following formula : ； in, represents the frequency of occurrence of amino acid a at position i'' in the sequence; represents the frequency of occurrence of amino acid b at position i'' in the sequence; A is a set of natural amino acids; The score of position i'' and amino acid a in the sequence is calculated according to the following formula : ； Where λ is a scaling factor used to adjust the scale of the score; q(a) is the average frequency of amino acid a in nature; The average score p[a] of amino acid a was calculated according to the following formula: ； Where N is the total length of the sequence.

5. The protein toxicity prediction method according to claim 1, characterized in that: The step of performing dimensionality reduction screening on the first feature vector to obtain a second feature vector of a target dimension includes: Delete the columns with a standard deviation of 0 in the first eigenvector to obtain the third eigenvector; Determine the Pearson correlation coefficient of each feature with the other features in the third feature vector; Randomly delete one of the two features whose Pearson correlation coefficient is greater than the correlation coefficient threshold to obtain the fourth eigenvector; constructing a co-selection relationship graph between features in the fourth feature vector to identify features associated with protein toxicity; The features with the largest weights of the target number are selected as representatives and passed to recursive feature elimination based on cross-validation to obtain the second feature vector of the target dimension.

6. A protein toxicity prediction system based on sequence information, characterized in that: include: A sequence acquisition module is used to acquire multiple protein sequences from a protein database; wherein the multiple protein sequences include toxic and non-toxic protein sequences; The feature calculation module is used to calculate the features of all protein sequences and obtain the feature vectors of six categories of proteins; A vector concatenation module is used to concatenate the six types of feature vectors by column to obtain the first feature vector; A dimension reduction screening module is used to perform dimension reduction screening on the first feature vector to obtain a second feature vector of a target dimension; A model training module, used for training a graph attention-based neural network model using the second eigenvector, and using the trained neural network model as a protein toxicity prediction model; The toxicity prediction module is used to input a new protein sequence into the protein toxicity prediction model to obtain a toxicity prediction result output by the protein toxicity prediction model.

7. The protein toxicity prediction system according to claim 6, characterized in that: The feature calculation module comprises: A first determining unit, used to determine a characteristic vector of protein R; A second determining unit is used to determine a feature vector of a position-specific scoring matrix based on evolutionary features; A third determining unit is used to determine a feature vector of an amino acid index based on physicochemical properties; A fourth determining unit, used to determine a characteristic vector of the frequency count of the amino acid group; a fifth determining unit, for determining a feature vector of a length of a protein sequence; The sixth determining unit is used to determine the feature vector of the protein common fragment sequence.

8. The protein toxicity prediction system according to claim 7, characterized in that: The first determining unit includes: The first construction means is used to construct the expression of the characteristic vector AAC of amino acid composition: ； Among them, p(a i ) is the ith amino acid a in the sequence i The frequency of occurrence of i ) is the ith amino acid a in the sequence i The number of occurrences; i=1,2,...,20; N is the total length of the sequence; And / or, a second constructing means for constructing an expression of a characteristic vector DPC of a dipeptide composition: ； Among them, p(d i,j ) is the dipeptide d in the sequence i,j The frequency of occurrence of f(d i,j ) is the dipeptide d in the sequence i,j The number of occurrences of; j=1,2,...,20; And / or, a third constructing means for constructing an expression for the characteristic vector MB of the Moreau-Broto autocorrelation: ； Among them, R d is the Moreau-Broto autocorrelation value of the sequence with delay d; d = 1, 2, ..., D; D represents the maximum delay; p i is the physicochemical property value of the ith amino acid in the sequence; p i+d is the physicochemical property value of the i+dth amino acid in the sequence; And / or, a fourth constructing means for constructing an expression of the characteristic vector MORAN of Moran autocorrelation: ； Among them, M d is the Moran autocorrelation value of the sequence with a delay of d; is the mean of all physicochemical property values; And / or, a fifth constructing means for constructing an expression for the characteristic vector GEARY of the Geary autocorrelation: ； Among them, G d is the Geary autocorrelation value of the sequence with a delay of d; And / or, a sixth constructing means for constructing an expression of a characteristic vector CTD of a combined transformation distribution: ； Among them, count(A i' ) is the i'th amino acid A i' The number of occurrences of Indicates the i'th amino acid A in the sequence i' and the j'th amino acid A j' The number of adjacencies; The representation captures the spatial distribution characteristics of each amino acid class; An operator representing the concatenation vector; And / or, a seventh constructing means for constructing an expression of a feature vector CTRIAD of a joint triple: ； in, Represents a triple The normalized frequency of occurrence of Represents a triple in a sequence The number of occurrences of g i is the group number of the ith amino acid in the sequence; g j is the group number of the jth amino acid in the sequence; is the group number of the k'th amino acid in the sequence; And / or, an eighth constructing means for constructing an expression of a characteristic vector SOCN of a sequence order coupling number: ； in, represents the number of sequential order couplings of delay d in the sequence on the physicochemical property k''; is the value of the physicochemical property k'' of the i-th amino acid; Representation and The amino acid property value of delay d; Represents the global variance of the attribute value, used for normalization; is the mean value of the physicochemical property k''; K is the total number of physicochemical properties; And / or, a ninth constructing means for constructing an expression of a characteristic vector QSO of a quasi-sequence order descriptor: ； in, It represents the coupling characteristics of the delay d in the sequence on the physicochemical property k''; And / or, a tenth construction means for constructing an expression of a characteristic vector PAAC of pseudo amino acid composition: ； in, represents the value of the physicochemical property k'' of the autocorrelation part of the sequence with delay d.

9. The protein toxicity prediction system according to claim 7, characterized in that: The second determining unit includes: The first calculation means is used to calculate the correction probability of amino acid a at position i'' in the sequence according to the following formula : ； in, represents the frequency of occurrence of amino acid a at position i'' in the sequence; represents the frequency of occurrence of amino acid b at position i'' in the sequence; A is a set of natural amino acids; The second calculation device is used to calculate the score of position i'' and amino acid a in the sequence according to the following formula : ； Where λ is a scaling factor used to adjust the scale of the score; q(a) is the average frequency of amino acid a in nature; The third calculating means is used to calculate the average score p[a] of amino acid a according to the following formula: ； Where N is the total length of the sequence.

10. The protein toxicity prediction system according to claim 6, characterized in that: The dimension reduction screening module includes: A first deletion unit is used to delete the columns with a standard deviation of 0 in the first eigenvector to obtain a third eigenvector; A determination unit, used to determine the Pearson correlation coefficient between each feature in the third feature vector and other features; The second deletion unit is used to randomly delete one of the two features whose Pearson correlation coefficient is greater than the correlation coefficient threshold, so as to obtain a fourth feature vector; a construction unit for constructing a co-selection relationship graph between features in the fourth feature vector to identify features associated with protein toxicity; The selection unit is used to select the target number of features with the largest weight as representatives, and pass them to the recursive feature elimination based on cross-validation to obtain the second feature vector of the target dimension.

Images