Pesticide target protein screening method and device based on entropy weight TOPSIS algorithm

Abstract

The invention discloses a pesticide target protein screening method and device based on an entropy weight TOPSIS algorithm, and the method comprises the steps: obtaining chemical structure characteristics of a to-be-screened pesticide, inputting the chemical structure characteristics into a plurality of databases, and predicting to obtain pesticide-protein affinity values corresponding to different databases; complementing and unifying protein information in pesticide-protein according to different databases, constructing a pesticide-protein affinity matrix, and performing normalization processing according to different databases to obtain a normalized pesticide-protein affinity matrix; and inputting the normalized pesticide-protein affinity matrix into an entropy weight TOPSIS algorithm to obtain a comprehensive evaluation index of each pesticide-protein, and screening out candidate target proteins based on the comprehensive evaluation indexes of the pesticide-protein, so as to solve the problem of high cost of pesticide target screening in ecotoxicology research.

Description

Technical Field

[0001] The present invention relates to the field of biological data processing, and in particular to a method and a device for screening pesticide target proteins based on an entropy weight TOPSIS algorithm. Background Art

[0002] As widely used chemicals, pesticides have potential impacts on human health and the ecological environment that cannot be ignored. Long-term chronic exposure to pesticides may have a variety of adverse effects on the environment and human health, including but not limited to reproductive toxicity, neurotoxicity, immunotoxicity, hepatotoxicity and nephrotoxicity. In addition, the environmental residue problem of pesticides, especially the use of per- and polyfluoroalkyl compounds (PFAS), has attracted widespread attention worldwide. Due to their persistence and bioaccumulation, these compounds may affect the health of humans and aquatic ecosystems through the food chain.

[0003] Network pharmacology is an emerging interdisciplinary subject that combines systems biology, computer science and pharmacology. It studies the multi-target mechanism of action of drugs by constructing a network relationship between drugs, targets and diseases. Unlike traditional single-target pharmacology, network pharmacology focuses on the analysis of the overall system and can fully reveal the mechanism of action and potential side effects of drugs in vivo. Through network pharmacology, we can better understand the multi-target effects of pesticides in organisms and evaluate their comprehensive effects on different biological systems. Using big data and bioinformatics technology, network pharmacology can screen out potentially harmful pesticides in a short time and provide a reliable theoretical basis for further experimental research. In the study of pesticide toxicity, network pharmacology has unique advantages. It can not only identify the effects of pesticides on specific targets, but also reveal the synergistic effects and potential toxic pathways of pesticides in different biological systems. This multi-level analysis method helps to comprehensively evaluate the long-term effects of pesticides on human health and the ecological environment and promote the safe use and management of pesticides.

[0004] At present, most of the mechanistic toxicology studies are conducted through model organisms such as zebrafish. Among them, the main methods of target evaluation and screening are based on molecular biology technology, layer by layer recursion or combined with a single target database for simple predictive analysis. These methods have problems such as high accuracy but high price or low price but low accuracy in determining pesticide targets. Therefore, there is an urgent need for a screening method that combines price and accuracy to assist in exploring the negative effects of chronic pesticide exposure on human health and contribute to achieving the goal of sustainable agricultural development. Summary of the invention

[0005] The purpose of this application is to propose a method and device for pesticide target protein screening based on entropy weight TOPSIS algorithm in response to the above-mentioned technical problems.

[0006] In a first aspect, the present invention provides a method for screening pesticide target proteins based on an entropy weight TOPSIS algorithm, comprising the following steps:

[0007] Obtain the chemical structure characteristics of the pesticides to be screened and input them into multiple databases respectively, and predict the pesticide-protein affinity values ​​corresponding to different databases. The chemical structure characteristics include 3D chemical structure formula and Canonical SMILE;

[0008] The protein information in pesticide-protein is supplemented and unified according to different databases, and the pesticide-protein affinity matrix is ​​constructed;

[0009] Normalizing the pesticide-protein affinity matrix according to different databases to obtain a normalized pesticide-protein affinity matrix;

[0010] The normalized pesticide-protein affinity matrix was input into the entropy weight TOPSIS algorithm to obtain the comprehensive evaluation index of each pesticide-protein, and the candidate target proteins were screened based on the comprehensive evaluation index of pesticide-protein.

[0011] Preferably, the plurality of databases include SwissTargetPrediction database, SuperPred database, PharmMapper database and SEA database.

[0012] As a preference, the protein information in the pesticide-protein is supplemented and unified according to different databases, and a pesticide-protein affinity matrix is ​​constructed, specifically including:

[0013] Convert the protein ID in the "Uniprot ID" column in the SuperPred database to the protein ID in Gene Symbol format;

[0014] Convert the protein numbers in the "Uniplot" column of the PharmMapper database to protein numbers in Gene Symbol format;

[0015] The protein numbers in the "Symbol" column in the SwissTargetPrediction database and the SEA database are the protein numbers in the Gene Symbol format;

[0016] The pesticide-protein affinity matrix X is expressed as:

[0017]

[0018] Where m is the total number of potential target proteins, n is the number of databases, and x is mnrepresents the affinity value of the mth target protein in the nth database, x ij represents the affinity value of the i-th target protein in the j-th database, i = 1, 2, ..., m, j = 1, 2, ..., n.

[0019] Preferably, the pesticide-protein affinity matrix is ​​normalized according to different databases to obtain a normalized pesticide-protein affinity matrix, which specifically includes:

[0020] Each element in the pesticide-protein affinity matrix is ​​normalized using the following formula to obtain the normalized element:

[0021]

[0022] Among them, z ij Represents x ij After normalization, we get the normalized elements. and They represent the minimum and maximum affinity values ​​of proteins in the jth database respectively;

[0023] The normalized pesticide-protein affinity matrix is ​​expressed as Z = [z ij ] m×n .

[0024] Preferably, the normalized pesticide-protein affinity matrix is ​​input into the entropy weight TOPSIS algorithm to obtain a comprehensive evaluation index for each pesticide-protein, specifically including:

[0025] The feature weight of each database was calculated based on the normalized pesticide-protein affinity matrix. As shown below:

[0026]

[0027] The Shannon entropy of the jth database is calculated using the maximum likelihood method based on the feature weight of each database. As shown below:

[0028]

[0029] The weight matrix W = [w1,…,w j ,…,w n ],j=1,2,…,n;

[0030]

[0031] in, is the Shannon entropy of the jth database, w jis the scoring weight of the j-th database;

[0032] The following formula is used for each score weight w in the weight matrix W j Perform an overall translation with a small positive increment ε to obtain the translated score weight w j ′ , ε>0:

[0033] w j ′ =w j +ε;

[0034] According to the translated score weight w j ′ Construct the weight matrix C = [c ij ] m×n , as shown below:

[0035]

[0036] Among them, c ij =w j ′ z ij ;

[0037] Calculate the positive ideal solution C according to the weighted matrix * and negative ideal solution C 0 , as shown below:

[0038] Assume that the positive ideal solution C * The attribute value of the jth database in is Negative ideal solution C 0 The attribute value of the jth database in is but:

[0039]

[0040]

[0041] The Euclidean distance is used to evaluate the distance of the pesticide-protein to the positive ideal solution or the negative ideal solution:

[0042]

[0043] in, represents the distance from the pesticide-protein to the positive ideal solution, Represents the distance from the pesticide-protein to the negative ideal solution

[0044] The comprehensive evaluation index of each pesticide-protein was calculated using the following formula, as shown below:

[0045]

[0046] Among them, the comprehensive evaluation index

[0047] Preferably, the candidate target proteins of the pesticides to be screened are screened based on the comprehensive evaluation index of pesticides-proteins, specifically including:

[0048] All the comprehensive evaluation indexes of pesticides-proteins are sorted from large to small, and the protein information in the pesticide-protein with the largest comprehensive evaluation index is selected as the candidate target protein of the pesticide.

[0049] In a second aspect, the present invention provides a pesticide target protein screening device based on an entropy weight TOPSIS algorithm, comprising:

[0050] The affinity value calculation module is configured to obtain the chemical structure characteristics of the pesticide to be screened and input them into multiple databases respectively, and predict the pesticide-protein affinity values ​​corresponding to different databases. The chemical structure characteristics include 3D chemical structure formula and Canonical SMILE;

[0051] A completion module is configured to complete and unify the protein information in the pesticide-protein according to different databases, and construct an affinity matrix of the pesticide-protein;

[0052] A normalization module is configured to normalize the pesticide-protein affinity matrix according to different databases to obtain a normalized pesticide-protein affinity matrix;

[0053] The index calculation module is configured to input the normalized pesticide-protein affinity matrix into the entropy weight TOPSIS algorithm to obtain the comprehensive evaluation index of each pesticide-protein, and screen out candidate target proteins based on the comprehensive evaluation index of pesticide-protein.

[0054] In a third aspect, the present invention provides an electronic device comprising one or more processors; a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0055] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method described in any implementation manner in the first aspect.

[0056] In a fifth aspect, the present invention provides a computer program product, comprising a computer program, which, when executed by a processor, implements the method described in any implementation manner in the first aspect.

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] (1) The pesticide target protein screening method based on entropy weight TOPSIS algorithm proposed in the present invention adopts network pharmacology method and applies PharmMapper, SwissTargetPrediction, SuperPred and SEA database prediction to preliminarily obtain the potential target protein information of pesticides.

[0059] (2) The pesticide target protein screening method based on the entropy weight TOPSIS algorithm proposed in the present invention uses the entropy weight TOPSIS algorithm to comprehensively evaluate the pesticide-protein affinity values ​​of multiple databases, and has good objectivity and robustness.

[0060] (3) The pesticide target protein screening method based on entropy weight TOPSIS algorithm proposed in the present invention solves the high cost problem of pesticide target screening in ecotoxicology research. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0062] Figure 1 A schematic diagram of a method for screening pesticide target proteins based on an entropy weight TOPSIS algorithm according to an embodiment of the present application;

[0063] Figure 2 A schematic diagram of the chemical structure characteristics of the pesticide target protein screening method based on the entropy weight TOPSIS algorithm of an embodiment of the present application;

[0064] Figure 3 A flowchart of an entropy weighted TOPSIS algorithm for a pesticide target protein screening method based on an entropy weighted TOPSIS algorithm according to an embodiment of the present application;

[0065] Figure 4 A comparison chart of the prediction performance evaluation indexes of the pesticide target protein screening method based on the entropy weight TOPSIS algorithm of the embodiment of the present application and the normalized sum and the entropy weight method-weighted sum;

[0066] Figure 5 Schematic diagram of a pesticide target protein screening device based on entropy weight TOPSIS algorithm according to an embodiment of the present application;

[0067] Figure 6 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all 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.

[0069] Figure 1 The present invention provides a method for screening pesticide target proteins based on an entropy weight TOPSIS algorithm, comprising the following steps:

[0070] S1, obtain the chemical structure characteristics of the pesticides to be screened and input them into multiple databases respectively, and predict the pesticide-protein affinity values ​​corresponding to different databases. The chemical structure characteristics include 3D chemical structure formula and CanonicalSMILE.

[0071] In a specific embodiment, the plurality of databases include a SwissTargetPrediction database, a SuperPred database, a PharmMapper database, and a SEA database.

[0072] Step 1. Specifically, we take the widely existing persistent pollutant perfluorooctanoic acid (PFOA) in my country as an example.

[0073] Taking PFOA as an example, the pesticide target protein screening method based on entropy weight TOPSIS algorithm proposed in the embodiment of the present application is described. First, the 3D chemical structure formula and Canonical SMILE of the pesticide to be screened are obtained, such as Figure 2 shown.

[0074] Secondly, based on the chemical structure characteristics of pesticides, multiple databases are combined for learning and prediction. The databases include

[0075] SwissTargetPrediction database, SuperPred database, PharmMapper database and SEA database.

[0076] The 3D chemical structure in sdf format and Canonical SMILE were input into the SwissTargetPrediction database (http: / / swisstargetprediction.ch / ), PharmMapper database (https: / / www.lilab-ecust.cn / pharmmapper / ), SEA database (https: / / sea.bkslab.org / ) and SuperPred database (https: / / prediction.charite.de / subpages / target_prediction.php), respectively. The species of the SwissTargetPrediction database was set to "Homo sapiens"; the "target setting" of the PharmMapper database was set to "Human Protein Targets Only (v2010,2241)". The final prediction results are shown in Tables 1-4, which include the affinity values ​​of pesticides and proteins.

[0077] Table 1

[0078]

[0079]

[0080] Table 2

[0081]

[0082] Table 3

[0083]

[0084] Table 4

[0085]

[0086] S2, the protein information in pesticide-protein is completed and unified according to different databases, and the pesticide-protein affinity matrix is ​​constructed.

[0087] In a specific embodiment, step S2 specifically includes:

[0088] Convert the protein ID in the "Uniprot ID" column in the SuperPred database to the protein ID in Gene Symbol format;

[0089] Convert the protein numbers in the "Uniplot" column of the PharmMapper database to protein numbers in Gene Symbol format;

[0090] The protein numbers in the "Symbol" column in the SwissTargetPrediction database and the SEA database are the protein numbers in the Gene Symbol format;

[0091] The pesticide-protein affinity matrix X is expressed as:

[0092]

[0093] Where m is the total number of potential target proteins, n is the number of databases, and x is mn represents the affinity value of the mth target protein in the nth database, x ij represents the affinity value of the i-th target protein in the j-th database, i = 1, 2, ..., m, j = 1, 2, ..., n.

[0094] Specifically, the protein information in the pesticide-protein was completed through the bioDBnet website (https: / / biodbnet-abcc.ncifcrf.gov / db / db2db.php). Using the "Uniplot" column in the prediction results of the PharmMapper database and the "Uniprot ID" column in the prediction results of the SuperPred database as templates, "Uniprot Accession" was selected as input and "Gene Symbol" was selected as output for conversion, and the protein number was unified into the GeneSymbol format, as shown in Table 5, where the protein information was taken from the corresponding database column or the result obtained by converting the corresponding database column, and the affinity value was taken from the value of the corresponding database column.

[0095] Table 5

[0096]

[0097] The unified protein information in different databases can be used with the corresponding affinity values ​​to construct the pesticide-protein affinity matrix, as shown in Table 6.

[0098] Table 6

[0099]

[0100] S3, normalizing the pesticide-protein affinity matrix according to different databases to obtain a normalized pesticide-protein affinity matrix.

[0101] In a specific embodiment, step S2 specifically includes:

[0102] Each element in the pesticide-protein affinity matrix is ​​normalized using the following formula to obtain the normalized element:

[0103]

[0104] Among them, z ij Represents x ij After normalization, we get the normalized elements. and They represent the minimum and maximum affinity values ​​of proteins in the jth database respectively;

[0105] The normalized pesticide-protein affinity matrix is ​​expressed as Z = [z ij ] m×n .

[0106] Specifically, the pesticide-protein affinity matrix X is dimensionless, i.e., normalized, according to the properties of the database. In order to avoid division by zero errors, when the maximum and minimum values ​​of the data in the jth column are equal, i.e. When , all the normalized values ​​of the data in this column are assigned to 1 to ensure robustness. The normalized pesticide-protein affinity matrix is ​​shown in Table 7.

[0107] Table 7

[0108]

[0109]

[0110] S4, input the normalized pesticide-protein affinity matrix into the entropy weight TOPSIS algorithm to obtain the comprehensive evaluation index of each pesticide-protein, and screen out candidate target proteins based on the comprehensive evaluation index of pesticide-protein.

[0111] In a specific embodiment, the normalized pesticide-protein affinity matrix is ​​input into the entropy weight TOPSIS algorithm to obtain a comprehensive evaluation index for each pesticide-protein, specifically including:

[0112] The feature weight of each database was calculated based on the normalized pesticide-protein affinity matrix. As shown below:

[0113]

[0114] The Shannon entropy of the jth database is calculated using the maximum likelihood method based on the feature weight of each database. As shown below:

[0115]

[0116] The weight matrix W = [w1,…,w j ,…,w n ],j=1,2,…,n;

[0117]

[0118] in, is the Shannon entropy of the jth database, w j is the scoring weight of the j-th database;

[0119] The following formula is used for each score weight w in the weight matrix W j Perform an overall translation with a small positive increment ε to obtain the translated score weight w j ′ , ε>0:

[0120] w j ′ =w j +ε;

[0121] According to the translated score weight w j ′ Construct the weight matrix C = [c ij ] m×n , as shown below:

[0122]

[0123] Among them, c ij =w j ′ z ij ;

[0124] Calculate the positive ideal solution C according to the weighted matrix * and negative ideal solution C 0 , as shown below:

[0125] Assume that the positive ideal solution C * The attribute value of the jth database in is Negative ideal solution C 0 The attribute value of the jth database in is but:

[0126]

[0127] The Euclidean distance is used to evaluate the distance of the pesticide-protein to the positive ideal solution or the negative ideal solution:

[0128]

[0129] in, represents the distance from the pesticide-protein to the positive ideal solution, Represents the distance from the pesticide-protein to the negative ideal solution

[0130] The comprehensive evaluation index of each pesticide-protein was calculated using the following formula, as shown below:

[0131]

[0132] Among them, the comprehensive evaluation index

[0133] In a specific embodiment, the candidate target proteins of the pesticides to be screened are screened based on the comprehensive evaluation index of pesticides-proteins, specifically including:

[0134] All the comprehensive evaluation indexes of pesticides-proteins are sorted from large to small, and the protein information in the pesticide-protein with the largest comprehensive evaluation index is selected as the candidate target protein of the pesticide.

[0135] The specific TOPSIS (Technique for Order Preference by Similarity to an Ideal Solution) model is a multi-attribute decision analysis method used to rank each solution by calculating the distance between each solution and the ideal solution and the negative ideal solution. The advantage of the TOPSIS model is that it can comprehensively consider multiple evaluation indicators, and the method is simple and computationally efficient, making it very suitable for comprehensive evaluation of complex systems. In the evaluation of pesticide toxicity, the TOPSIS model can help objectively analyze the comprehensive toxicity of different pesticides and find the optimal or suboptimal pesticide use plan.

[0136] The entropy weight method is a commonly used weight determination method. Based on the information entropy of each indicator, the importance of each indicator is objectively determined. The information entropy reflects the degree of discreteness of the indicator. The larger the entropy value, the greater the variability of the indicator, and the higher the weight in the comprehensive evaluation. Combining the entropy weight method and the TOPSIS model, the embodiments of the present application can more objectively evaluate the toxicity and environmental impact of pesticides, reduce the influence of subjective factors on the results, and improve the scientificity and reliability of the evaluation results. In specific applications, the weights of each toxicity index are first calculated by the entropy weight method, and then the TOPSIS model is used to comprehensively score and rank different pesticides, providing a method that is both objective, robust and generalizable for predicting the toxicity targets of pesticides.

[0137] Specifically, refer to Figure 3, the normalized pesticide-protein affinity matrix is ​​used as input, and the entropy weight method is used to assign scoring weights to the normalized pesticide-protein affinity values ​​of different databases. Weights are assigned to different databases based on the discreteness and information entropy of the scores of each database. First, the feature weight of the database is calculated. Secondly, the Shannon entropy of the jth database is calculated using the maximum likelihood method. The higher the entropy value of the Shannon entropy, the lower the purity of the information and the more information it contains; the smaller the entropy value of the Shannon entropy, the higher the purity of the information and the less information it contains. Since the affinity values ​​between databases are independently distributed, the entropy value of the Shannon entropy calculated here is the empirical entropy based on the maximum likelihood method. Then, according to the entropy value of the Shannon entropy Calculate the weight matrix. Since each element in the weight matrix must be positive in the process of solving the TOPSIS model, in order to avoid the scoring weight w j When w is 0, j The overall translation is performed with a small positive increment ε to improve the generalization of the model. The score weights after translation are shown in Table 8, which ultimately form the translated weight matrix W′.

[0138] Table 8

[0139]

[0140] Finally, combined with the translated scoring weights obtained by the entropy weight method, the TOPSIS method is used to calculate a comprehensive evaluation index for pesticide-protein. The TOPSIS method calculates the distance between each object and the positive ideal solution and the negative ideal solution to obtain the relative proximity as the comprehensive scoring index. Specifically, the weight matrix C is first constructed according to the translated weight matrix W′, and the positive ideal solution C is determined by the weight matrix C. * and negative ideal solution C 0 . is the attribute value of the jth database in the positive ideal solution, that is, the maximum value that database j can reach in the weighted matrix C; is the attribute value of the jth database in the negative ideal solution, that is, the minimum value that database j can achieve in the weighted matrix C. The attribute values ​​of all databases belong to benefit-type attributes. Furthermore, based on the positive ideal solution and the negative ideal solution, the Euclidean distance is used to evaluate the affinity of pesticides and proteins, thereby calculating the comprehensive evaluation index of pesticides and proteins. The larger the value, the greater the affinity of the pesticide-protein pair. The results are shown in Table 9.

[0141] Table 9

[0142]

[0143]

[0144] Comprehensive evaluation index based on TOPSIS method The pesticide-proteins were ranked, and those with higher scores were the candidate target proteins of the pesticides. The results are shown in Table 10.

[0145] Table 10

[0146]

[0147] The results showed that protein NR1H2 was the key target protein most closely related to the toxic effects of the pesticide PFOA. Its distance to the negative ideal solution was 1.120058576, its distance to the positive ideal solution was 1.446192065, and its comprehensive evaluation index was 0.580808637.

[0148] According to the pesticide target protein screening method based on the entropy weight TOPSIS algorithm proposed in the embodiment of the present application, the candidate target proteins of PFOA are calculated, and then the reliability of the results and the robustness of the model are verified by using methods such as literature research and molecular docking.

[0149] 1) Indicator selection

[0150] The above prediction results are divided into true positive (TP, True Positive), false positive (FP, False Positive) and false negative (FN, False Negative). Among them:

[0151] ① True Positive (TP): The number of instances correctly predicted as positive by the prediction results of the embodiments of the present application, i.e., proteins that are actually pesticide targets and are also predicted by the model to be target proteins of pesticides.

[0152] ②True Negative (TN): The number of instances correctly predicted as negative in the prediction results of the embodiments of the present application, that is, the target protein is not actually a pesticide, and the model also predicts it as a non-pesticide target protein. Due to the large number of true negative targets, their impact on the final result will be excluded.

[0153] ③ False Positive (FP): The number of instances in which the prediction results of the embodiments of the present application are incorrectly predicted as positive, that is, they are not actually target proteins of pesticides, but are incorrectly predicted as target proteins of pesticides by the model.

[0154] ④ False Negative (FN): The number of instances in which the prediction results of the embodiments of the present application are incorrectly predicted as negative classes, that is, proteins that are actually pesticide targets but are incorrectly predicted by the model as non-pesticide targets.

[0155] Based on the classification results, the accuracy, precision, recall rate and F1 score are selected for evaluation. The above indicators are calculated by classifying the prediction results of the pesticide target protein screening method based on the entropy weight TOPSIS algorithm proposed in the embodiment of the present application, providing a multi-dimensional performance evaluation.

[0156] 2) Literature research

[0157] Furthermore, based on the literature survey, the number of false negative (FN) targets was determined by consulting the Pubmed and Embase databases, i.e., the human protein molecular targets of PFOA obtained through experimental verification. The experimental subjects included in the literature only include human tissue cells or human body. The list of true negative targets is shown in Table 11.

[0158] Table 11

[0159]

[0160] 3) Molecular docking simulation

[0161] In order to prove the robustness and objectivity of the pesticide target protein screening method based on the entropy weight TOPSIS algorithm proposed in the embodiment of the present application, the binding activity between the top 15 proteins screened by the three methods and the pesticide PFOA was detected by molecular docking simulation. First, the target protein was searched in the Uniprot database (https: / / www.uniprot.org / ), the protein number corresponding to the target protein was obtained, the protein number was searched in the RCSB PDB database and the pdb file of the protein was downloaded. Then the target protein was hydrogenated and dehydrated in the Schrondinger software, the binding position of the ligand and the receptor was found and the binding energy was calculated. The calculation results are shown in Table 12. If the binding energy is lower than -5.0 kcal / mol, it means that its binding performance is good, and the larger the absolute value, the more stable the conformation.

[0162] Table 12

[0163]

[0164] 4) Results

[0165] As shown in Table 13, the entropy weighted TOPSIS algorithm has better performance in identifying true positives and false positives, while it is as effective as the normalized sum method in identifying false negatives.

[0166] Table 13

[0167]

[0168] Furthermore, the accuracy, precision, recall and F1-score are calculated based on the following formulas. The results are shown in Table 14:

[0169]

[0170] Table 14

[0171]

[0172] Combined with Table 14 and Figure 4 It can be seen that all indicators of the entropy weight TOPSIS algorithm exceed the normalized sum and the entropy weight method-weighted sum; Figure 4 The horizontal axis is different indicators, and the vertical axis is the evaluation index value. Due to its objectivity and robustness, the entropy weight TOPSIS algorithm can be well used in the prediction and screening of target proteins of pesticides.

[0173] The present invention can solve the technical bottleneck of the existing research that makes it difficult to explore the molecular mechanism of pesticides, provide key data support for public health monitoring, and help to detect potential health risks early. At the same time, it can optimize the pesticide use strategy in agricultural production, reduce environmental residues, promote the development of precision agriculture, ensure crop yield and quality, and achieve sustainable agricultural goals.

[0174] Further references Figure 5 As an implementation of the methods shown in the above figures, the present application provides an embodiment of a pesticide target protein screening device based on an entropy weight TOPSIS algorithm. Figure 1 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0175] The present application embodiment provides a pesticide target protein screening device based on entropy weight TOPSIS algorithm, comprising:

[0176] Affinity value calculation module 1 is configured to obtain chemical structure characteristics of the pesticide to be screened and input them into multiple databases respectively, and predict the pesticide-protein affinity values ​​corresponding to different databases, the chemical structure characteristics including 3D chemical structure formula and Canonical SMILE;

[0177] The completion module 2 is configured to complete and unify the protein information in the pesticide-protein according to different databases, and construct an affinity matrix of the pesticide-protein;

[0178] A normalization module 3 is configured to normalize the pesticide-protein affinity matrix according to different databases to obtain a normalized pesticide-protein affinity matrix;

[0179] The index calculation module 4 is configured to input the normalized pesticide-protein affinity matrix into the entropy weight TOPSIS algorithm to obtain the comprehensive evaluation index of each pesticide-protein, and screen out candidate target proteins based on the comprehensive evaluation index of the pesticide-protein.

[0180] Figure 6 Schematic diagram of the hardware structure of the electronic device provided by the embodiment of the present invention. Figure 6 As shown, the electronic device of this embodiment includes: a processor 601 and a memory 602; wherein the memory 602 is used to store computer-executable instructions; the processor 601 is used to execute the computer-executable instructions stored in the memory to implement the various steps performed by the electronic device in the above embodiment. For details, please refer to the relevant description in the above method embodiment.

[0181] Optionally, the memory 602 may be independent or integrated with the processor 601 .

[0182] When the memory 602 is independently provided, the electronic device further includes a bus 603 for connecting the memory 602 and the processor 601 .

[0183] The embodiment of the present invention further provides a computer storage medium, in which computer execution instructions are stored. When the processor 601 executes the computer execution instructions, the above method is implemented.

[0184] The embodiment of the present invention further provides a computer program product, including a computer program. When the computer program is executed by the processor 601, the above method is implemented.

[0185] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0186] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to implement the solution of this embodiment.

[0187] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, or each module may exist physically separately, or two or more modules may be integrated into one unit. The above-mentioned module-composed unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0188] The above-mentioned integrated module implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor 601 to perform some steps of the methods of various embodiments of the present application.

[0189] It should be understood that the processor 601 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or the processor 601 may be any conventional processor 601, etc. The steps of the method disclosed in the invention may be directly embodied in the execution of the hardware processor 601, or may be executed by a combination of hardware and software modules in the processor 601.

[0190] The memory 602 may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk memory, and may also be a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk, or an optical disk.

[0191] The bus 603 may be an Industry Standard Architecture (ISA), a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus 603 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus 603 in the drawings of the present application is not limited to only one bus 603 or one type of bus 603.

[0192] The above storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0193] An exemplary storage medium is coupled to the processor 601, so that the processor 601 can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor 601. The processor 601 and the storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor 601 and the storage medium can also exist as discrete components in an electronic device or a main control device.

[0194] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0195] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for screening pesticide target proteins based on entropy weight TOPSIS algorithm, characterized in that: The following steps are involved: Obtain the chemical structure characteristics of the pesticides to be screened and input them into multiple databases respectively, and predict the pesticide-protein affinity values ​​corresponding to different databases. The chemical structure characteristics include 3D chemical structure formula and Canonical SMILE; The protein information in the pesticide-protein is supplemented and unified according to different databases, and an affinity matrix of the pesticide-protein is constructed; Normalizing the pesticide-protein affinity matrix according to different databases to obtain a normalized pesticide-protein affinity matrix; The normalized pesticide-protein affinity matrix is ​​input into the entropy weight TOPSIS algorithm to obtain a comprehensive evaluation index of each pesticide-protein, and candidate target proteins are screened out based on the comprehensive evaluation index of the pesticide-protein.

2. The method for screening pesticide target proteins based on entropy weight TOPSIS algorithm according to claim 1, characterized in that: The multiple databases include SwissTargetPrediction database, SuperPred database, PharmMapper database and SEA database.

3. The method for screening pesticide target proteins based on entropy weight TOPSIS algorithm according to claim 2, characterized in that: The protein information in the pesticide-protein is supplemented and unified according to different databases, and an affinity matrix of the pesticide-protein is constructed, specifically including: Convert the protein number in the "Uniprot ID" column in the SuperPred database into a protein number in Gene Symbol format; Convert the protein number in the "Uniplot" column in the PharmMapper database into the protein number in Gene Symbol format; The protein numbers in the "Symbol" column in the SwissTargetPrediction database and the SEA database are protein numbers in the Gene Symbol format; The pesticide-protein affinity matrix X is expressed as: Where m is the total number of potential target proteins, n is the number of databases, and x is mn represents the affinity value of the mth target protein in the nth database, x ij represents the affinity value of the i-th target protein in the j-th database, i = 1, 2, ..., m, j = 1, 2, ..., n.

4. The method for screening pesticide target proteins based on entropy weight TOPSIS algorithm according to claim 3, characterized in that: The pesticide-protein affinity matrix is ​​normalized according to different databases to obtain a normalized pesticide-protein affinity matrix, which specifically includes: Each element in the pesticide-protein affinity matrix is ​​normalized using the following formula to obtain a normalized element: Among them, z ij Represents x ij After normalization, we get the normalized elements. and They represent the minimum and maximum affinity values ​​of proteins in the jth database respectively; The normalized pesticide-protein affinity matrix is ​​expressed as Z = [z ij ] m×n .

5. The method for screening pesticide target proteins based on entropy weight TOPSIS algorithm according to claim 1, characterized in that: The normalized pesticide-protein affinity matrix was input into the entropy weight TOPSIS algorithm to obtain the comprehensive evaluation index of each pesticide-protein, including: The feature weight of each database is calculated based on the normalized pesticide-protein affinity matrix. As shown below: The Shannon entropy of the jth database is calculated using the maximum likelihood method according to the feature weight of each database. As shown below: The weight matrix W=[w1,…,w j ,…,w n ],j=1,2,…,n; in, is the Shannon entropy of the jth database, w j is the scoring weight of the j-th database; The following formula is used for each score weight w in the weight matrix W j Perform an overall translation with a small positive increment ε to obtain the translated score weight w j ′ , ε>0: w j ′ =w j +ε; According to the translated scoring weight w j ′ Construct the weighted matrix C = [c ij ] m×n , as shown below: Among them, c ij =w j ′ z ij ; Calculate the positive ideal solution C according to the weighting matrix * and negative ideal solution C 0 , as shown below: Assume that the positive ideal solution C * The attribute value of the jth database in is Negative ideal solution C 0 The attribute value of the jth database in is but: The Euclidean distance is used to evaluate the distance of the pesticide-protein to the positive ideal solution or the negative ideal solution: in, represents the distance from the pesticide-protein to the positive ideal solution, Represents the distance from the pesticide-protein to the negative ideal solution The comprehensive evaluation index of each pesticide-protein was calculated using the following formula, as shown below: Among them, the comprehensive evaluation index 6. The method for screening pesticide target proteins based on entropy weight TOPSIS algorithm according to claim 1, characterized in that: Screening out the candidate target protein of the pesticide to be screened based on the comprehensive evaluation index of the pesticide-protein specifically includes: All the comprehensive evaluation indexes of pesticides-proteins are sorted from large to small, and the protein information in the pesticide-protein with the largest comprehensive evaluation index is selected as the candidate target protein of the pesticide.

7. A pesticide target protein screening device based on entropy weight TOPSIS algorithm, characterized in that: include: The affinity value calculation module is configured to obtain the chemical structure characteristics of the pesticide to be screened and input them into multiple databases respectively, and predict the pesticide-protein affinity values ​​corresponding to different databases. The chemical structure characteristics include 3D chemical structure formula and Canonical SMILE; A completion module is configured to complete and unify the protein information in the pesticide-protein according to different databases, and construct an affinity matrix of the pesticide-protein; A normalization module is configured to normalize the pesticide-protein affinity matrix according to different databases to obtain a normalized pesticide-protein affinity matrix; The index calculation module is configured to input the normalized pesticide-protein affinity matrix into the entropy weight TOPSIS algorithm to obtain the comprehensive evaluation index of each pesticide-protein, and screen out candidate target proteins based on the comprehensive evaluation index of the pesticide-protein.

8. An electronic device, comprising: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images