Method, device, storage medium and electronic equipment for acquiring key similar pollutants

Abstract

The application relates to a method, device, storage medium and electronic equipment for obtaining a key similar pollutant, the method comprising: obtaining a simplified molecular linear input specification file corresponding to a target CAS number of a target pollutant in a target environment and a target pollutant descriptor set of the target pollutant comprising a plurality of target descriptors from a pre-acquired literature database according to the target CAS number; calculating the similarity between the target pollutant and other pollutants in the literature database based on each similarity algorithm and the target pollutant descriptor set; obtaining similar pollutants similar to the target pollutant based on the similarity and a similarity threshold value corresponding to each similarity algorithm; obtaining a prediction parameter of the target pollutant and the similar pollutants according to a K nearest neighbor algorithm; obtaining a pharmacokinetic parameter of the pollutant in a living body based on the prediction parameter and a PBTK model; and evaluating the target environment based on the pharmacokinetic parameter. The efficiency of obtaining the key similar pollutant can be improved.

Description

Technical Field

[0001] This invention relates to the field of ecological monitoring technology, and in particular to a method, apparatus, storage medium, and electronic device for obtaining key similar pollutants. Background Technology

[0002] In the wave of global industrialization, the application of chemicals in human production systems and daily life scenarios is becoming increasingly widespread. For example, from catalysts used to accelerate reactions on factory assembly lines, pesticides used to protect agricultural harvests, additives used to enhance the texture of cosmetics, and special solvents used to maintain the performance of electronic devices, tens of thousands of chemicals are being integrated into every aspect of society on an astonishing scale.

[0003] Among these chemicals, persistent organic pollutants (POPs) are emerging pollutants that pose a threat to ecology and human health as "invisible killers." POPs possess strong biotoxicity, interfering with normal physiological functions. They are difficult to degrade under natural conditions and can persist in the environment for decades or even longer. Through bioaccumulation, they can accumulate along the food chain, eventually reaching dangerous concentrations in higher organisms. This poses a significant potential risk to the growth and reproduction of plants and animals, ecosystem stability, and even the nervous and reproductive systems of humans. Furthermore, this risk often has a long-term latency, causing irreversible damage to ecosystems. Therefore, based on the key pollutants in the target environment, similar key pollutants are obtained. A Physiologically Based Pharmacokinetics (PBTK) model of the key pollutants in the target environment within the human body is then established based on these key pollutants and similar key pollutants. The absorption, distribution, metabolism, and excretion processes of these key pollutants in the human body are then analyzed using the constructed PBTK model, thereby assessing the health risks of the target environment and providing a scientific basis for toxicological research and environmental health risk assessment of the target environment.

[0004] Biotoxicokinetic models, based on knowledge of physiology, biochemistry, and anatomy, simulate blood flow in the body's circulatory system, connecting various tissues and organs, and following the principle of mass balance to simulate the dynamic processes of chemical substances within the body. This provides a scientific basis for toxicological research and environmental health risk assessment. However, this method requires identifying similar key pollutants that play a crucial role in the target environment using biotoxicity testing. This process is time-consuming and costly, making it crucial to improve the efficiency of obtaining these similar pollutants a pressing technical challenge. Summary of the Invention

[0005] In view of this, the present invention provides a method, apparatus, storage medium and electronic device for obtaining key similar pollutants.

[0006] Specifically, the present invention is achieved through the following technical solution:

[0007] According to a first aspect of the present invention, a method for obtaining key similar pollutants is provided, the method comprising:

[0008] Based on the target CAS number of the target pollutant in the target environment, obtain the simplified molecular linear input specification file corresponding to the target CAS number from the pre-acquired literature database;

[0009] The target pollutant descriptor set is obtained based on the simplified molecular linear input specification file, and the target pollutant descriptor set includes multiple target descriptors;

[0010] Based on the pre-set similarity algorithms and the target pollutant descriptor set, the similarity between the target pollutant and other pollutants in the literature database is calculated respectively. Based on the similarity and the similarity thresholds corresponding to the pre-set similarity algorithms, similar pollutants similar to the target pollutant are obtained.

[0011] Based on the K-nearest neighbor algorithm, prediction parameters of the target pollutant and similar pollutants are obtained. Based on the prediction parameters and the pre-built PBTK model, the pharmacokinetic parameters of the pollutants in the organism are obtained, so as to assess the target environment based on the pharmacokinetic parameters.

[0012] Optionally, obtaining the simplified molecular linear input specification file corresponding to the target CAS number from a pre-acquired literature database includes:

[0013] Call the strsplit function in the literature database to split the input target CAS number string and obtain individual CAS number units;

[0014] The strtrim function in the literature database is called to process the individual CAS number units and obtain a list of standardized CAS numbers;

[0015] The getSMILESFromCAS helper function is called to obtain the corresponding SMILES file based on the standard CAS number list by interacting with the chemical database in the literature repository.

[0016] Optionally, obtaining the target pollutant descriptor set based on the simplified molecular linear input specification file includes:

[0017] Based on the number of molecular structures corresponding to each simplified molecular linear input specification in the simplified molecular linear input specification file, the pre-set PaDEL descriptor calculation function is called to obtain the molecular descriptor set corresponding to the target pollutant.

[0018] Optionally, the step of calculating the similarity between the target pollutant and other pollutants in the literature database based on pre-set similarity algorithms and the target pollutant descriptor set includes:

[0019] The target pollutant descriptor set is filtered according to a pre-set filtering strategy to obtain a filtered descriptor set.

[0020] Based on the filtered descriptor set, the similarity between the filtered descriptor set and the pollutant descriptor sets of other pollutants in the literature database is calculated using pre-set similarity algorithms.

[0021] Optionally, the step of filtering the acquired target pollutant descriptor set according to a pre-set filtering strategy to obtain filtered descriptors includes:

[0022] According to the pre-set filtering rules, the obtained target pollutant descriptor set is filtered to obtain the first candidate filtering descriptor. The filtering rules are used to remove missing values, outliers and redundant values ​​from the target pollutant descriptor set.

[0023] Statistical analysis is performed on each candidate descriptor in the first candidate screening descriptor set to obtain descriptors that meet the pre-set pollutant property conditions, thus obtaining the screening descriptor set.

[0024] Optionally, the step of calculating the similarity between the selected descriptor set and the pollutant descriptor sets of other pollutants in the literature database using pre-set similarity algorithms includes:

[0025] Traverse each filter descriptor in the filter descriptor set, and for the pollutant descriptor set of the first pollutant in the literature database, obtain the pollutant descriptor to be calculated that is the same as the filter descriptor;

[0026] Using the Euclidean distance algorithm, the first similarity between the target pollutant and the first pollutant is obtained based on the straight-line distance between the filter descriptor and the pollutant descriptor to be calculated;

[0027] Using the Manhattan distance algorithm, the second similarity between the target pollutant and the first pollutant is obtained based on the sum of the absolute differences between the filter descriptors and the pollutant descriptors to be calculated;

[0028] Using the cosine similarity algorithm, the third similarity between the target pollutant and the first pollutant is obtained based on the cosine value of the angle between the filtered descriptor and the pollutant descriptor to be calculated.

[0029] Using the Tanimoto coefficient algorithm, the fourth similarity between the target pollutant and the first pollutant is obtained based on the ratio of the intersection and union of the screening descriptors and the pollutant descriptors to be calculated.

[0030] Optionally, the step of obtaining similar pollutants similar to the target pollutant based on similarity and pre-set similarity thresholds corresponding to each similarity algorithm includes:

[0031] For each other pollutant, if the first similarity is less than a pre-set Euclidean distance threshold, the second similarity is less than a pre-set Manhattan threshold, the third similarity is greater than a pre-set cosine similarity threshold, and the fourth similarity is greater than a pre-set Tanimoto coefficient threshold, then the other pollutant is determined to be a similar pollutant to the target pollutant.

[0032] The method for obtaining key similar pollutants in this technical solution involves: obtaining a simplified molecular linear input specification file corresponding to the target CAS number of the target pollutant in the target environment from a pre-acquired literature database; obtaining a target pollutant descriptor set based on the simplified molecular linear input specification file, the target pollutant descriptor set including multiple target descriptors; calculating the similarity between the target pollutant and other pollutants in the literature database based on pre-set similarity algorithms and the target pollutant descriptor set; obtaining similar pollutants similar to the target pollutant based on the similarity and the similarity thresholds corresponding to the pre-set similarity algorithms; obtaining prediction parameters of the target pollutant and similar pollutants based on the K-nearest neighbor algorithm; and obtaining the pharmacokinetic parameters of the pollutants in vivo based on the prediction parameters and a pre-constructed PBTK model, so as to evaluate the target environment based on the pharmacokinetic parameters. In this way, by parsing the CAS number, the target pollutant descriptor set is obtained, and similarity analysis using multiple similarity algorithms is performed to obtain similar pollutants that are similar to the target pollutant. Then, prediction is performed based on the K nearest neighbor algorithm. Based on the prediction parameters and the pre-built PBTK model, the target environment is evaluated, which can accurately screen key similar pollutants, thereby improving the efficiency of obtaining key similar pollutants.

[0033] According to a second aspect of the present invention, an apparatus for obtaining key similar pollutants is provided, the apparatus comprising:

[0034] The file acquisition module is used to obtain the simplified molecular linear input specification file corresponding to the target CAS number from a pre-acquired literature database based on the target CAS number of the target pollutant in the target environment.

[0035] The descriptor acquisition module is used to acquire a target pollutant descriptor set based on a simplified molecular linear input specification file, wherein the target pollutant descriptor set includes multiple target descriptors;

[0036] The pollutant acquisition module is used to calculate the similarity between the target pollutant and other pollutants in the literature database based on pre-set similarity algorithms and the target pollutant descriptor set, and to acquire similar pollutants similar to the target pollutant based on the similarity and the similarity thresholds corresponding to the pre-set similarity algorithms.

[0037] The parameter evaluation module is used to obtain the predicted parameters of the target pollutant and similar pollutants based on the K-nearest neighbor algorithm. Based on the predicted parameters and the pre-built PBTK model, the pharmacokinetic parameters of the pollutants in the organism are obtained, so as to evaluate the target environment based on the pharmacokinetic parameters.

[0038] According to a third aspect of the invention, a storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method for obtaining key similar contaminants in any possible implementation of the first aspect.

[0039] According to a fourth aspect of the present invention, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method for obtaining key similar contaminants in any possible implementation of the first aspect. Attached Figure Description

[0040] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0042] Figure 1 A flowchart illustrating a method for obtaining key similar pollutants provided in an embodiment of the present invention;

[0043] Figure 2 A schematic diagram of an apparatus for obtaining key similar pollutants provided in an embodiment of the present invention;

[0044] Figure 3 Another schematic diagram of an apparatus for obtaining key similar pollutants provided in an embodiment of the present invention;

[0045] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] See Figure 1 This invention provides a method for obtaining key similar pollutants, which may include the following steps:

[0048] S101. Based on the target CAS number of the target pollutant in the target environment, obtain the simplified molecular linear input specification file corresponding to the target CAS number from the pre-acquired literature database;

[0049] In this embodiment, the Simplified Molecular InputLine Entry System (SMILES) file is a specification file that explicitly describes the molecular structure of pollutants using ASCII strings. As an optional embodiment, the library pre-stores the correspondence between CAS numbers and SMILES files characterizing molecular structure information. In the library, each organic pollutant molecule has a corresponding CAS number, which is a unique identifier for the organic pollutant.

[0050] In this embodiment, as an optional implementation, a simplified molecular linear input specification file corresponding to the target CAS number is obtained from a pre-acquired literature database, including:

[0051] Call the strsplit function in the literature database to split the input target CAS number string and obtain individual CAS number units;

[0052] The strtrim function in the literature database is called to process the individual CAS number units and obtain a list of standardized CAS numbers;

[0053] The getSMILESFromCAS helper function is called to obtain the corresponding SMILES file based on the standard CAS number list by interacting with the chemical database in the literature repository.

[0054] In this embodiment, the user can input multiple CAS number strings corresponding to target pollutants, separated by commas. For example, if the user inputs the CAS number string "67-66-3, 1163-19-5", the strsplit function will split it by commas, resulting in two independent units representing the CAS number: "67-66-3" and "1163-19-5". Following the standardized format of CAS numbers, the strtrim function will remove leading and trailing spaces, resulting in a standardized list of CAS numbers: "67-66-3" and "1163-19-5". Finally, by calling the getSMILESFromCAS helper function, the system interacts with the chemical database to retrieve the corresponding SMILES string for each CAS number in the CAS number list. The results are displayed in a multi-line text area (SMILESDisplayTextArea), with one SMILES string displayed in each text area for easy viewing, copying, or further use by the user.

[0055] S102. Obtain the target pollutant descriptor set based on the simplified molecular linear input specification file, wherein the target pollutant descriptor set includes multiple target descriptors;

[0056] In this embodiment, the descriptors include, but are not limited to, molecular descriptors and molecular fingerprints. A molecular descriptor is a quantitative indicator reflecting the physicochemical properties of the target pollutant molecule, while a molecular fingerprint is a code string describing the structural characteristics of a molecule.

[0057] In this embodiment, as an optional implementation, obtaining the target pollutant descriptor set based on the simplified molecular linear input specification file includes:

[0058] Based on the number of molecular structures corresponding to each simplified molecular linear input specification in the simplified molecular linear input specification file, the pre-set PaDEL descriptor calculation function is called to obtain the molecular descriptor set corresponding to the target pollutant.

[0059] In this embodiment, the simplified molecular linear input specification file includes multiple simplified molecular linear input specifications, and the pollutant contains multiple pollutant molecules, with each pollutant molecule corresponding to a simplified molecular linear input specification. As an optional embodiment, the `ComputeButtonPushed` function is called to obtain from the simplified molecular linear input specification file the number of molecular structures (numStruct), the path of the smi structure file (StructureFile) containing the simplified molecular linear input specification, whether to calculate 2D descriptors (D_2d), whether to calculate 3D descriptors (D_3d), and the fingerprint type to be obtained (D_fp_select). The `Calculating_PaDEL_descriptors` function in the PaDEL descriptor calculation function is then called to read the molecular structures in the smi file according to the structure file path, filter target molecules according to the set number of molecular structures (numStruct), calculate the corresponding molecular descriptors based on the selection status of D_2d and D_3d, and extract the specified type of molecular fingerprint according to the fingerprint type to be obtained.

[0060] In this embodiment, as an optional embodiment, the method further includes:

[0061] The acquired target pollutant descriptor set is organized and visualized.

[0062] In this embodiment, the target pollutant descriptor set includes multiple key indicator features, each corresponding to a descriptor. As an optional embodiment, the path to the calculation result text file is obtained from the interface, the file content is read and converted into a table format, and simultaneously saved as a MAT format file for later retrieval. For example, subsequently, by reloading the saved MAT file, it is verified whether it contains valid table data: if valid, the molecule name ("Name" column) and nine key fingerprint features ("FP1" to "FP9" columns), totaling ten columns of core data, are extracted and displayed in a table on the interface for easy and intuitive viewing by the user.

[0063] S103. Based on the pre-set similarity algorithms and the target pollutant descriptor set, calculate the similarity between the target pollutant and other pollutants in the literature database. Based on the similarity and the similarity thresholds corresponding to the pre-set similarity algorithms, obtain similar pollutants that are similar to the target pollutant.

[0064] In this embodiment, based on the target pollutant descriptor, the similarity between the target pollutant and other pollutants in the literature database is calculated using multiple similarity algorithms to obtain similar pollutants.

[0065] In this embodiment, as an optional implementation, based on pre-set similarity algorithms and the target pollutant descriptor set, the similarity between the target pollutant and other pollutants in the literature database is calculated, including:

[0066] A11. Based on the pre-set screening strategy, the acquired target pollutant descriptor set is screened to obtain the screened descriptor set;

[0067] In this embodiment, the calculated molecular descriptor set (target pollutant descriptor set) is screened to remove invalid information and retain key features. As an optional embodiment, the acquired target pollutant descriptor set is screened according to a pre-set screening strategy to obtain screened descriptors, including:

[0068] According to the pre-set filtering rules, the obtained target pollutant descriptor set is filtered to obtain the first candidate filtering descriptor. The filtering rules are used to remove missing values, outliers and redundant values ​​from the target pollutant descriptor set.

[0069] In this embodiment, the screening rules are used to remove missing values ​​(no calculation results), outliers (values ​​that deviate significantly from the reasonable range), and redundant values ​​(highly repetitive and indistinguishable descriptors, such as the descriptor values ​​being completely identical for all molecules).

[0070] Statistical analysis is performed on each candidate descriptor in the first candidate screening descriptor set to obtain descriptors that meet the pre-set pollutant property conditions, thus obtaining the screening descriptor set.

[0071] In this embodiment, statistical analysis includes, but is not limited to, analysis of variance and correlation analysis. Through statistical analysis, descriptors that have a significant impact on pollutant properties (such as toxicity and mobility) and high discriminative power (e.g., descriptors directly related to pollutant characteristics such as molecular weight and solubility) are retained. For example, candidate descriptors whose toxicity impact exceeds a pre-set toxicity threshold are selected, ultimately resulting in a set of effective molecular descriptors (screened descriptors). As an optional embodiment, the set of effective molecular descriptors may contain multiple specific descriptors. For example, the set of effective molecular descriptors for a pollutant may include dozens or even hundreds of specific descriptors such as molecular weight, solubility, and octanol-water partition coefficient. Each descriptor reflects the physicochemical properties of the pollutant from different dimensions.

[0072] A12, based on the filtered descriptor set, uses pre-set similarity algorithms to calculate the similarity between the filtered descriptor set and the pollutant descriptor sets of other pollutants in the literature database.

[0073] In this embodiment, as an optional implementation, the similarity algorithm includes, but is not limited to, Euclidean distance, Manhattan distance, cosine similarity, and the Tanimoto coefficient. Therefore, using the pre-set similarity algorithms, the similarity between the selected descriptor set and the pollutant descriptor sets of other pollutants in the literature database is calculated, including:

[0074] Traverse each filter descriptor in the filter descriptor set, and for the pollutant descriptor set of the first pollutant in the literature database, obtain the pollutant descriptor to be calculated that is the same as the filter descriptor;

[0075] Using the Euclidean distance algorithm, the first similarity between the target pollutant and the first pollutant is obtained based on the straight-line distance between the filter descriptor and the pollutant descriptor to be calculated;

[0076] Using the Manhattan distance algorithm, the second similarity between the target pollutant and the first pollutant is obtained based on the sum of the absolute differences between the filter descriptors and the pollutant descriptors to be calculated;

[0077] Using the cosine similarity algorithm, the third similarity between the target pollutant and the first pollutant is obtained based on the cosine value of the angle between the filtered descriptor and the pollutant descriptor to be calculated.

[0078] Using the Tanimoto coefficient algorithm, the fourth similarity between the target pollutant and the first pollutant is obtained based on the ratio of the intersection and union of the screening descriptors and the pollutant descriptors to be calculated.

[0079] In this embodiment, Euclidean Distance is used to calculate the straight-line distance between two molecular descriptor vectors to measure similarity. The specific formula is as follows:

[0080]

[0081] Where X and Y are the first molecule descriptor and the second molecule descriptor, respectively, xi and yi are the i-th descriptors of the first molecule and the second molecule, respectively, and n is the number of descriptors.

[0082] Manhattan Distance is used to calculate the sum of the absolute differences between the vectors corresponding to two molecular descriptors to measure similarity. The specific formula is as follows:

[0083] )

[0084] Cosine Similarity is used to calculate the cosine of the angle between two molecular descriptor vectors to measure their similarity. The specific formula is as follows:

[0085] )

[0086] The Tanimoto Coefficient is used to calculate the ratio of the intersection to the union of two vectors (or sets) to measure similarity. In this embodiment, for the vectors X=(x1,x2,…,xn) and Y=(y1,y2,…,yn) corresponding to the first and second molecule descriptors, respectively, the specific formula is as follows:

[0087] ).

[0088] In this embodiment, the effective molecular descriptor set of the target pollutant is compared with the effective molecular descriptor set of each pollutant in the database one by one using the above four similarity calculations. Similarity is not determined by any single similarity method; instead, a similarity threshold is preset for each of the four similarity algorithms. For example, for the cosine similarity algorithm, a cosine similarity ≥ 0.8 is set, i.e., the cosine similarity threshold is set to 0.8; for the Euclidean distance algorithm, an Euclidean distance ≤ 2 is set, i.e., the Euclidean distance threshold is set to 2. Only pollutants that meet the preset thresholds are considered similar to the target pollutant, and thus the CAS number corresponding to the similar pollutant is obtained.

[0089] In this embodiment, as an optional implementation, similar pollutants similar to the target pollutant are obtained based on similarity and the similarity thresholds corresponding to each pre-set similarity algorithm, including:

[0090] For each other pollutant, if the first similarity is less than a pre-set Euclidean distance threshold, the second similarity is less than a pre-set Manhattan threshold, the third similarity is greater than a pre-set cosine similarity threshold, and the fourth similarity is greater than a pre-set Tanimoto coefficient threshold, then the other pollutant is determined to be a similar pollutant to the target pollutant.

[0091] In this embodiment, after calculation using four similarity algorithms, each candidate pollutant (target pollutant) yields four similarity-related values ​​(two distance values ​​and two similarity values). The similarity results for multiple pollutants can be sorted. However, the sorting is not a weighted sum of the four values; instead, it is sorted separately according to the similarity algorithm type. If the similarity result is cosine similarity or Tanimoto coefficient, it is sorted in descending order of value; that is, the closer the similarity result value is to 1, the higher the ranking. If the similarity result is Euclidean distance or Manhattan distance, it is sorted in ascending order of value; that is, the closer the value is to 0, the higher the ranking.

[0092] In this embodiment, as an optional embodiment, after sorting is completed, the sorting results (including the CAS numbers of similar pollutants and the corresponding similarity values) are saved to the target file.

[0093] In this embodiment, as another optional embodiment, the method further includes:

[0094] The pre-built pollutant database is updated based on the pollutants corresponding to the calculated similarity.

[0095] In this embodiment, a target file is constructed based on the pollutants corresponding to the calculated similarity. The pollutants in the target file are compared and analyzed with a pre-constructed pollutant database. As an optional embodiment, this includes:

[0096] The calculated similarity values ​​are normalized to obtain normalized values.

[0097] In this embodiment, since the similarity values ​​obtained by the four similarity algorithms are of different magnitudes, for example, the distance value may be 0-10 and the similarity value may be 0-1, all similarity values ​​are normalized to map the values ​​of different ranges to the interval [0, 1], so that the results of each similarity algorithm can be directly compared.

[0098] Extract known similar pollutants mapped from the pollutant database, and calculate the overlap between the candidate similar pollutants corresponding to the target pollutant determined by the similarity algorithm and the known similar pollutants.

[0099] In this embodiment, matching accuracy is verified by extracting known similar pollutants of the target pollutant from the pollutant database and comparing the overlap between the calculated similar pollutants (candidate similar pollutants) and the known similar pollutants. As an optional embodiment, a higher overlap ratio indicates a smaller deviation between the calculated similarity value and the similarity value recorded in the pollutant database, resulting in higher matching accuracy. As another optional embodiment, the proportions of "false positives" (calculated as similar but actually dissimilar) and "false negatives" (actually similar but not calculated) are also statistically analyzed to quantify accuracy.

[0100] If the overlap is greater than a preset overlap threshold, based on the candidate similar pollutants corresponding to the target pollutant determined by the similarity algorithm, if the pollutant database does not store the same similar pollutant as the candidate similar pollutant, the candidate similar pollutant is placed in the similar pollutant character library corresponding to the target pollutant in the pollutant database.

[0101] In this embodiment, database coverage verification can be performed: the proportion of candidate similar pollutants that can be found in the pollutant database is calculated. The higher the proportion, the higher the coverage of the pollutant database for that type of pollutant. If there are similar pollutants that cannot be found in the pollutant database, they are recorded as uncovered pollutants, and the pollutant database is updated to provide a basis for subsequent pollutant database updates.

[0102] In this embodiment, as an optional implementation, the verification results are output in tabular form to intuitively display the correlation between the target pollutant and similar pollutants and the analysis results. The verification results include, but are not limited to: the CAS number of the target pollutant, the CAS numbers of similar pollutants, the normalized similarity values ​​after each similarity algorithm, the matching accuracy index, and the coverage index.

[0103] S104. Based on the K-nearest neighbor algorithm, obtain the predicted parameters of the target pollutant and similar pollutants. Based on the predicted parameters and the pre-built PBTK model, obtain the pharmacokinetic parameters of the pollutants in the organism, so as to assess the target environment based on the pharmacokinetic parameters.

[0104] In this embodiment, the K-Nearest Neighbors (KNN) algorithm is used to predict based on similar pollutants. The prediction parameters include, but are not limited to, pharmacokinetic core metabolic parameters, including, but not limited to, CLINT parameters, FUB parameters, Clint parameters, and Caco2 parameters.

[0105] In this embodiment, the K-nearest neighbor algorithm is a quantitative structure-activity relationship (QSAR) method. The QSAR method is used to calculate the core metabolic parameters of drugs for similar pollutants that are similar to the target pollutant, and subsequent predictions are made based on the core metabolic parameters of drugs.

[0106] In this embodiment, as an optional implementation, a QSAR model can be trained based on a large amount of experimental data to calculate the core metabolic parameters of drugs for similar pollutants to the target pollutant. Specifically, based on descriptors such as the lipid solubility, polarity, and metabolic sites of each pollutant molecule, the intrinsic liver clearance rate is obtained by fitting the QSAR model, i.e., CLINT = a × lipid solubility + b × polarity + c × number of metabolic sites + ..., where a, b, and c are coefficients obtained from training the QSAR model, and CLINT reflects the rate at which the liver metabolizes the pollutant; the larger the value, the faster the liver metabolizes the pollutant.

[0107] In this embodiment, the unbound plasma fraction (FuB) is fitted based on descriptors such as the hydrophilicity / hydrophobicity of each pollutant molecule and the number of plasma protein binding sites, i.e., FuB = d × hydrophilicity / hydrophobicity + e × number of binding sites + ..., where d and e are coefficients. FuB reflects the proportion of pollutants in the blood that are not bound to plasma proteins; the higher the value, the more pollutants are free in the blood and can exert their effects. For the permeability of human colon adenocarcinoma cells (CACO2), it is fitted based on descriptors such as molecular weight and lipid-water partition coefficient, i.e., CACO2 = f × molecular weight + g × lipid-water partition coefficient + ..., where f and g are coefficients. CACO2 reflects the ability of pollutants to pass through intestinal epithelial cells; the higher the value, the easier it is for the pollutants to be absorbed into the bloodstream from the intestine.

[0108] In this embodiment, as an optional implementation, pharmacokinetic parameters can be used to determine the volume of distribution in vivo, elimination half-life, and concentration distribution in various organs. Pharmacokinetic parameters reflect the patterns of absorption, distribution, metabolism, and excretion of pollutants after they enter the body.

[0109] In this embodiment, the PBPK structural model, constructed by combining the characteristics of material dynamics, uses the chemical properties of the material and experimental research data to quantitatively describe the absorption, distribution, metabolism and excretion of the substance in the body through the mass conservation equation.

[0110] In this embodiment, as an optional embodiment, the method further includes:

[0111] The distribution of pollutant concentrations is displayed using line graphs, concentration distribution maps, and heat maps.

[0112] In this embodiment, the simulation results of the PBTK model are visualized and analyzed, including drawing line graphs, generating heat maps, and viewing and exporting images. This intuitively presents the concentration change patterns of pollutants in organisms, thereby enabling the formulation of corresponding technical solutions to improve the target environment based on the concentration change patterns.

[0113] In this embodiment, when drawing a line chart, the working path can be obtained from the interface, the path of the Excel file to be read (out_solve_pbtk.xlsx) can be recorded in the log, and in the try block, the data of the Excel file can be read through the readtable function, the data of the first column can be extracted as the x-axis (time) data, the data of columns 3-10 can be extracted as the concentration data of each organ and converted into a numerical vector, the organ node selected by the user in the tree control (Tree2_2) can be obtained, and the concentration-time line of the corresponding organ can be drawn in a loop: the concentration data column index can be matched by the organ name, the corresponding column data can be extracted, the plot function can be called to draw the line in the UIAxes control, and the legend name can be recorded simultaneously.

[0114] When generating and displaying the heatmap, the `plotAndSaveCorrelationHeatmap` function is called in the `try` block to generate the heatmap and save it as the `correlation_heatmap.png` file under the `savePath` path. The successfully saved heatmap image is read and displayed in the `uiaxHeatmap` control using the `imshow` function. The heatmap generation completion information is recorded in the log.

[0115] In this embodiment, when exporting an image, the handle of the UIAxes control is obtained, the path of the image export folder is read from the interface, and the existence of the path is checked. If the path does not exist, it is created automatically. The exportgraphics function is used to export the chart in UIAxes as a PNG file with a resolution of 300 (file name is exported_chart.png).

[0116] In this embodiment, organ concentration display can also be performed. Based on the simulation results of the PBTK model, pollutant concentration data of each organ at a specified time point is extracted. Through multi-mode display and targeted analysis, the distribution characteristics of pollutants in each organ are presented intuitively. For example, multi-organ concentration data calculation and multi-mode display can be performed by calling the calculation function get_points (reading the simulation results of the PBTK model, extracting the concentration data of a specified time point and a specified organ and calculating key evaluation indicators) or get_pointall (multi-organ batch analysis function), and passing in the working path, time point parameters, a predefined array of 7 organ names (lung, liver, kidney, vein, artery, plasma, intestine, etc.) and the display mode parameters selected by the user.

[0117] In this embodiment, multi-mode result display and interface synchronization updates are also possible: multi-mode data conversion provides four display modes to adapt to different analysis needs:

[0118] (1) p1 mode: Calculate the ratio of the current concentration to the maximum concentration (currentConcentrationSums / maxConcentrationSum) to reflect the relative distribution;

[0119] (2) p2 mode: Display normalized value (simplified to 1 here, retaining the interface for extended complex logic);

[0120] (3) Max mode: Displays the maximum concentration value for all organs;

[0121] (4) Current mode: Directly displays the absolute concentration value at the current time point.

[0122] In this embodiment, an integrated interface update can also be performed: the system updates the converted numerical results to the seven corresponding organ display controls (including lung, liver, kidney, vein, artery, plasma, and intestine), completing the entire process from data calculation to visualization. The system can also retrieve the name of the user-selected organ from the organDropDown dropdown menu to achieve organ-oriented analysis. For example, it can call the dedicated calculation function get_point to perform pharmacokinetic calculations and return four key results: current concentration value (currentConcentrationSum), maximum concentration value (maxConcentrationSum), statistical significance index (result_p), and image storage path (imagePath).

[0123] In this embodiment, a quantitative structure-activity relationship (QSAR) model characterizing the absorption, distribution, metabolism, and excretion processes in organisms is constructed using machine learning. Combined with high-throughput toxicology technology, a PBTK model is established to provide a visualization tool for chemical risk assessment, new drug development, and drug research in humans.

[0124] Based on the same inventive concept, such as Figure 2 As shown, embodiments of the present invention also provide an apparatus for obtaining key similar pollutants, the apparatus comprising:

[0125] The file acquisition module 201 is used to acquire the simplified molecular linear input specification file corresponding to the target CAS number from a pre-acquired literature database based on the target CAS number of the target pollutant in the target environment.

[0126] In this embodiment, as an optional embodiment, the file acquisition module 201 is specifically used for:

[0127] Call the strsplit function in the literature database to split the input target CAS number string and obtain individual CAS number units;

[0128] The strtrim function in the literature database is called to process the individual CAS number units and obtain a list of standardized CAS numbers;

[0129] The getSMILESFromCAS helper function is called to obtain the corresponding SMILES file based on the standard CAS number list by interacting with the chemical database in the literature repository.

[0130] The descriptor acquisition module 202 is used to acquire a target pollutant descriptor set for the target pollutant based on a simplified molecular linear input specification file, wherein the target pollutant descriptor set includes multiple target descriptors;

[0131] In this embodiment, the descriptor includes, but is not limited to, molecular descriptor and molecular fingerprint.

[0132] In this embodiment, as an optional embodiment, the descriptor acquisition module 202 is specifically used for:

[0133] Based on the number of molecular structures corresponding to each simplified molecular linear input specification in the simplified molecular linear input specification file, the pre-set PaDEL descriptor calculation function is called to obtain the molecular descriptor set corresponding to the target pollutant.

[0134] In this embodiment, as another optional embodiment, the descriptor acquisition module 202 is further configured to:

[0135] The acquired target pollutant descriptor set is organized and visualized.

[0136] The pollutant acquisition module 203 is used to calculate the similarity between the target pollutant and other pollutants in the literature database based on pre-set similarity algorithms and the target pollutant descriptor set, and to acquire similar pollutants similar to the target pollutant based on the similarity and the similarity thresholds corresponding to the pre-set similarity algorithms.

[0137] In this embodiment, as an optional embodiment, the pollutant acquisition module 203 is specifically used for:

[0138] The target pollutant descriptor set is filtered according to a pre-set filtering strategy to obtain a filtered descriptor set.

[0139] Based on the filtered descriptor set, the similarity between the filtered descriptor set and the pollutant descriptor sets of other pollutants in the literature database is calculated using pre-set similarity algorithms.

[0140] In this embodiment, as an optional embodiment, the acquired target pollutant descriptor set is filtered according to a pre-set filtering strategy to obtain filtered descriptors, including:

[0141] According to the pre-set filtering rules, the obtained target pollutant descriptor set is filtered to obtain the first candidate filtering descriptor. The filtering rules are used to remove missing values, outliers and redundant values ​​from the target pollutant descriptor set.

[0142] Statistical analysis is performed on each candidate descriptor in the first candidate screening descriptor set to obtain descriptors that meet the pre-set pollutant property conditions, thus obtaining the screening descriptor set.

[0143] In this embodiment, as an optional implementation, pre-set similarity algorithms are used to calculate the similarity between the selected descriptor set and the pollutant descriptor sets of other pollutants in the literature database, including:

[0144] Traverse each filter descriptor in the filter descriptor set, and for the pollutant descriptor set of the first pollutant in the literature database, obtain the pollutant descriptor to be calculated that is the same as the filter descriptor;

[0145] Using the Euclidean distance algorithm, the first similarity between the target pollutant and the first pollutant is obtained based on the straight-line distance between the filter descriptor and the pollutant descriptor to be calculated;

[0146] Using the Manhattan distance algorithm, the second similarity between the target pollutant and the first pollutant is obtained based on the sum of the absolute differences between the filter descriptors and the pollutant descriptors to be calculated;

[0147] Using the cosine similarity algorithm, the third similarity between the target pollutant and the first pollutant is obtained based on the cosine value of the angle between the filtered descriptor and the pollutant descriptor to be calculated.

[0148] Using the Tanimoto coefficient algorithm, the fourth similarity between the target pollutant and the first pollutant is obtained based on the ratio of the intersection and union of the screening descriptors and the pollutant descriptors to be calculated.

[0149] In this embodiment, as an optional implementation, similar pollutants similar to the target pollutant are obtained based on similarity and the similarity thresholds corresponding to each pre-set similarity algorithm, including:

[0150] For each other pollutant, if the first similarity is less than a pre-set Euclidean distance threshold, the second similarity is less than a pre-set Manhattan threshold, the third similarity is greater than a pre-set cosine similarity threshold, and the fourth similarity is greater than a pre-set Tanimoto coefficient threshold, then the other pollutant is determined to be a similar pollutant to the target pollutant.

[0151] In this embodiment, as another optional embodiment, the pollutant acquisition module 203 is further used for:

[0152] The pre-built pollutant database is updated based on the pollutants corresponding to the calculated similarity.

[0153] In this embodiment, as an optional implementation, a target file is constructed based on the pollutants corresponding to the calculated similarity. The pollutants in the target file are compared and analyzed with a pre-constructed pollutant database, including:

[0154] The calculated similarity values ​​are normalized to obtain normalized values.

[0155] Extract known similar pollutants mapped from the pollutant database, and calculate the overlap between the candidate similar pollutants corresponding to the target pollutant determined by the similarity algorithm and the known similar pollutants.

[0156] If the overlap is greater than a preset overlap threshold, based on the candidate similar pollutants corresponding to the target pollutant determined by the similarity algorithm, if the pollutant database does not store the same similar pollutant as the candidate similar pollutant, the candidate similar pollutant is placed in the similar pollutant character library corresponding to the target pollutant in the pollutant database.

[0157] The parameter evaluation module 204 is used to obtain the predicted parameters of the target pollutant and similar pollutants based on the K nearest neighbor algorithm, and to obtain the pharmacokinetic parameters of the pollutants in the organism based on the predicted parameters and the pre-built PBTK model, so as to evaluate the target environment based on the pharmacokinetic parameters.

[0158] In this embodiment, as an optional implementation, a QSAR model can be trained based on a large amount of experimental data to calculate the core metabolic parameters of drugs for similar pollutants to the target pollutant. Specifically, based on descriptors such as the lipid solubility, polarity, and metabolic sites of each pollutant molecule, the intrinsic liver clearance rate is obtained by fitting the QSAR model, i.e., CLINT = a × lipid solubility + b × polarity + c × number of metabolic sites + ..., where a, b, and c are coefficients obtained from training the QSAR model, and CLINT reflects the rate at which the liver metabolizes the pollutant; the larger the value, the faster the liver metabolizes the pollutant.

[0159] In this embodiment, as an optional embodiment, the parameter evaluation module 204 is further used for:

[0160] The distribution of pollutant concentrations is displayed using line graphs, concentration distribution maps, and heat maps.

[0161] Figure 3 This is another schematic diagram of an apparatus for obtaining key similar pollutants provided in an embodiment of the present invention. (See diagram below.) Figure 3 As shown, the device for obtaining key similar pollutants includes: a CAS parsing and SMILES generation module, a molecular descriptor calculation module, a descriptor screening module, a similarity screening module, a similarity verification module, a QSAR parameter prediction module, and a PBTK simulation and visualization module.

[0162] The CAS parsing and SMILES generation module is used to obtain the simplified molecular linear input specification file corresponding to the target CAS number from a pre-acquired literature database based on the target CAS number of the target pollutant in the target environment.

[0163] In this embodiment, the input CAS number, for example 108-88-3, is parsed to generate and output a SMILES file.

[0164] The molecular descriptor calculation module is used to obtain a target pollutant descriptor set based on a simplified molecular linear input specification file, wherein the target pollutant descriptor set includes multiple target descriptors;

[0165] In this embodiment, as an optional embodiment, the target descriptor is a molecular descriptor, and the calculation results are displayed. The calculation results include, but are not limited to, molecular weight, solubility, and LogP.

[0166] The descriptor filtering module is used to filter the acquired target pollutant descriptor set according to a pre-set filtering strategy to obtain a filtered descriptor set.

[0167] In this embodiment, the obtained filter descriptor set is the valid descriptor set.

[0168] The similarity filtering module is used to calculate the similarity between the target pollutant and other pollutants in the literature database based on pre-set similarity algorithms and the filtering descriptor set.

[0169] In this embodiment, as an optional implementation, the similarity algorithm includes, but is not limited to, Euclidean distance, Manhattan distance, and cosine similarity. Correspondingly, the calculation result display interface shows the similarity algorithm used and the similar pollutants obtained based on that similarity algorithm.

[0170] The similarity verification module is used to obtain similar pollutants that are similar to the target pollutant based on similarity and the similarity thresholds corresponding to each pre-set similarity algorithm.

[0171] In this embodiment, similarity verification is performed using the matching rate. As an optional embodiment, known similar pollutants mapped to the target pollutant are extracted from the pollutant database, and the overlap between the candidate similar pollutants corresponding to the target pollutant determined by the similarity algorithm and the known similar pollutants is calculated. If the overlap is greater than a preset overlap threshold, based on the candidate similar pollutants corresponding to the target pollutant determined by the similarity algorithm, if the pollutant database does not store the same similar pollutant as the candidate similar pollutant, the candidate similar pollutant is placed in the similar pollutant character set corresponding to the target pollutant in the pollutant database.

[0172] The QSAR parameter prediction module is used to obtain the prediction parameters of the target pollutant and similar pollutants based on the K nearest neighbor algorithm.

[0173] In this embodiment, the prediction results of core metabolic parameters are displayed on the interface.

[0174] The PBTK simulation and visualization module is used to obtain the pharmacokinetic parameters of pollutants in organisms based on predicted parameters and pre-built PBTK models, so as to assess the target environment based on the pharmacokinetic parameters.

[0175] Based on the same inventive concept, embodiments of the present invention also provide a storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method for obtaining key similar pollutants in any of the above possible implementations.

[0176] Optionally, the storage medium may be a non-transitory computer-readable storage medium, such as a ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device.

[0177] Based on the same inventive concept, see [link to inventive concept] Figure 4 This invention also provides an electronic device, including a memory 101 (e.g., non-volatile memory), a processor 102, and a computer program stored on the memory 101 and executable on the processor 102. When the processor 102 executes the program, it implements the steps of the method for obtaining key similar pollutants in any of the above possible implementations, which can be equivalent to the aforementioned apparatus for obtaining key similar pollutants. Of course, the processor can also be used to process other data or perform calculations. This electronic device can be a PC, server, terminal, or other similar device.

[0178] like Figure 4 As shown, the electronic device may also include: memory 103, network interface 104, and internal bus 105. In addition to these components, other hardware may also be included, which will not be described in detail here.

[0179] It should be noted that the above-mentioned device for obtaining key similar pollutants can be implemented by software. As a device in a logical sense, it is formed by the processor 102 of the electronic device in which it is located reading the computer program instructions stored in the non-volatile memory into the memory 103 for execution.

[0180] The embodiments of the subject matter and functional operation described in this specification can be implemented in the following ways: digital electronic circuits, tangibly embodied computer software or firmware, computer hardware including the structures disclosed in this specification and their structural equivalents, or combinations thereof. Embodiments of the subject matter described in this specification can be implemented as one or more computer programs, i.e., one or more modules of computer program instructions encoded on a tangible, non-transitory program carrier for execution by a data processing apparatus or for controlling the operation of a data processing apparatus. Alternatively or additionally, the program instructions may be encoded on artificially generated propagation signals, such as machine-generated electrical, optical, or electromagnetic signals, which are generated to encode information and transmit it to a suitable receiving device for execution by the data processing apparatus. The computer storage medium may be a machine-readable storage device, a machine-readable storage substrate, a random or serial access memory device, or combinations thereof.

[0181] The processing and logic flow described in this specification can be executed by one or more programmable computers that execute one or more computer programs to perform corresponding functions by operating on input data and generating output. The processing and logic flow can also be executed by special-purpose logic circuitry—such as FPGA (Field Programmable Gate Array) or ASIC (Application-Specific Integrated Circuit), and the device can also be implemented as special-purpose logic circuitry.

[0182] Suitable computers for executing computer programs include, for example, general-purpose and / or special-purpose microprocessors, or any other type of central processing unit. Typically, the central processing unit receives instructions and data from read-only memory and / or random access memory. The basic components of a computer include a central processing unit for implementing or executing instructions and one or more memory devices for storing instructions and data. Typically, a computer will also include one or more mass storage devices for storing data, such as disks, magneto-optical disks, or optical disks, or the computer will be operatively coupled to such mass storage devices to receive data from or transfer data to them, or both. However, a computer is not required to have such devices. Furthermore, a computer can be embedded in another device, such as a mobile phone, a personal digital assistant (PDA), a mobile audio or video player, a game console, a global positioning system (GPS) receiver, or a portable storage device such as a universal serial bus (USB) flash drive, to name a few.

[0183] Computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media, and memory devices, such as semiconductor memory devices (e.g., EPROM, EEPROM, and flash memory devices), magnetic disks (e.g., internal hard disks or removable disks), magneto-optical disks, and CD-ROM and DVD-ROM disks. Processors and memory may be supplemented by or incorporated into dedicated logic circuitry.

[0184] While this specification contains numerous specific implementation details, these should not be construed as limiting the scope of any invention or the scope of the claims, but rather are primarily used to describe features of specific embodiments of a particular invention. Certain features described in the various embodiments herein may also be implemented in combination in a single embodiment. Conversely, various features described in a single embodiment may also be implemented separately in various embodiments or in any suitable sub-combination. Furthermore, while features may function in certain combinations as described above and even initially claimed in this way, one or more features from a claimed combination may be removed from that combination in some cases, and a claimed combination may refer to a sub-combination or a variation of a sub-combination.

[0185] Similarly, although the operations are depicted in a specific order in the accompanying drawings, this should not be construed as requiring these operations to be performed in the specific order shown or sequentially, or requiring all illustrated operations to be performed to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous. Furthermore, the separation of various system modules and components in the above embodiments should not be construed as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0186] Thus, specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. In some cases, the actions recited in the claims may be performed in a different order and still achieve the desired result. Furthermore, the processes depicted in the drawings are not necessarily shown in a specific order or sequence to achieve the desired result. In some implementations, multitasking and parallel processing may be advantageous.

[0187] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0188] The above are merely specific embodiments of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for obtaining key similar pollutants, characterized in that, include: Based on the target CAS number of the target pollutant in the target environment, obtain the simplified molecular linear input specification file corresponding to the target CAS number from the pre-acquired literature database; The target pollutant descriptor set is obtained based on the simplified molecular linear input specification file, and the target pollutant descriptor set includes multiple target descriptors; Based on the pre-set similarity algorithms and the target pollutant descriptor set, the similarity between the target pollutant and other pollutants in the literature database is calculated respectively. Based on the similarity and the similarity thresholds corresponding to the pre-set similarity algorithms, similar pollutants similar to the target pollutant are obtained. Based on the K-nearest neighbor algorithm, prediction parameters of the target pollutant and similar pollutants are obtained. Based on the prediction parameters and the pre-built PBTK model, the pharmacokinetic parameters of the pollutants in the organism are obtained, so as to assess the target environment based on the pharmacokinetic parameters.

2. The method for obtaining key similar pollutants according to claim 1, characterized in that, The step of obtaining the simplified molecular linear input specification file corresponding to the target CAS number from a pre-acquired literature database includes: Call the strsplit function in the literature database to split the input target CAS number string and obtain individual CAS number units; The strtrim function in the literature database is called to process the individual CAS number units and obtain a list of standardized CAS numbers; The getSMILESFromCAS helper function is called to obtain the corresponding SMILES file based on the standard CAS number list by interacting with the chemical database in the literature repository.

3. The method for obtaining key similar pollutants according to claim 1, characterized in that, The process of obtaining the target pollutant descriptor set based on the simplified molecular linear input canonical file includes: Based on the number of molecular structures corresponding to each simplified molecular linear input specification in the simplified molecular linear input specification file, the pre-set PaDEL descriptor calculation function is called to obtain the molecular descriptor set corresponding to the target pollutant.

4. The method for obtaining key similar pollutants according to any one of claims 1 to 3, characterized in that, The step of calculating the similarity between the target pollutant and other pollutants in the literature database based on pre-set similarity algorithms and the target pollutant descriptor set includes: The target pollutant descriptor set is filtered according to a pre-set filtering strategy to obtain a filtered descriptor set. Based on the filtered descriptor set, the similarity between the filtered descriptor set and the pollutant descriptor sets of other pollutants in the literature database is calculated using pre-set similarity algorithms.

5. The method for obtaining key similar pollutants according to claim 4, characterized in that, The step of filtering the acquired target pollutant descriptor set according to a pre-set filtering strategy to obtain filtered descriptors includes: According to the pre-set filtering rules, the obtained target pollutant descriptor set is filtered to obtain the first candidate filtering descriptor. The filtering rules are used to remove missing values, outliers and redundant values ​​from the target pollutant descriptor set. Statistical analysis is performed on each candidate descriptor in the first candidate screening descriptor set to obtain descriptors that meet the pre-set pollutant property conditions, thus obtaining the screening descriptor set.

6. The method for obtaining key similar pollutants according to claim 4, characterized in that, The step of using pre-set similarity algorithms to calculate the similarity between the selected descriptor set and the pollutant descriptor sets of other pollutants in the literature database includes: Traverse each filter descriptor in the filter descriptor set, and for the pollutant descriptor set of the first pollutant in the literature database, obtain the pollutant descriptor to be calculated that is the same as the filter descriptor; Using the Euclidean distance algorithm, the first similarity between the target pollutant and the first pollutant is obtained based on the straight-line distance between the filter descriptor and the pollutant descriptor to be calculated; Using the Manhattan distance algorithm, the second similarity between the target pollutant and the first pollutant is obtained based on the sum of the absolute differences between the filter descriptors and the pollutant descriptors to be calculated; Using the cosine similarity algorithm, the third similarity between the target pollutant and the first pollutant is obtained based on the cosine value of the angle between the filtered descriptor and the pollutant descriptor to be calculated. Using the Tanimoto coefficient algorithm, the fourth similarity between the target pollutant and the first pollutant is obtained based on the ratio of the intersection and union of the screening descriptors and the pollutant descriptors to be calculated.

7. The method for obtaining key similar pollutants according to claim 4, characterized in that, The method, based on similarity and pre-set similarity thresholds corresponding to various similarity algorithms, obtains similar pollutants to the target pollutant, including: For each other pollutant, if the first similarity is less than a pre-set Euclidean distance threshold, the second similarity is less than a pre-set Manhattan threshold, the third similarity is greater than a pre-set cosine similarity threshold, and the fourth similarity is greater than a pre-set Tanimoto coefficient threshold, then the other pollutant is determined to be a similar pollutant to the target pollutant.

8. An apparatus for obtaining key similar pollutants, characterized in that, The apparatus for obtaining key similar pollutants includes: The file acquisition module is used to obtain the simplified molecular linear input specification file corresponding to the target CAS number from a pre-acquired literature database based on the target CAS number of the target pollutant in the target environment. The descriptor acquisition module is used to acquire a target pollutant descriptor set based on a simplified molecular linear input specification file, wherein the target pollutant descriptor set includes multiple target descriptors; The pollutant acquisition module is used to calculate the similarity between the target pollutant and other pollutants in the literature database based on pre-set similarity algorithms and the target pollutant descriptor set, and to acquire similar pollutants similar to the target pollutant based on the similarity and the similarity thresholds corresponding to the pre-set similarity algorithms. The parameter evaluation module is used to obtain the predicted parameters of the target pollutant and similar pollutants based on the K-nearest neighbor algorithm. Based on the predicted parameters and the pre-built PBTK model, the pharmacokinetic parameters of the pollutants in the organism are obtained, so as to evaluate the target environment based on the pharmacokinetic parameters.

9. A storage medium, characterized in that, A program or instructions are stored on a storage medium, and the program or instructions are executed by a processor to implement the steps of the method for obtaining key similar contaminants as described in any one of claims 1 to 7.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the method for obtaining key similar pollutants as described in any one of claims 1 to 7.

Images