Priority control OPEs screening method based on surface water concentration data

Abstract

This application relates to the technical field of environmental monitoring and data analysis, and particularly to a screening method for priority-controlled OPEs based on surface water concentration data. According to the organophosphate ester flame retardants (OPEs) exposure database, the monitoring data of each OPE compound within a specified time period is obtained. For each OPE compound, based on the monitoring data, the quantitative data of environmental persistence, bioaccumulation, and toxicity effects are respectively obtained, and the exposure potential index and hazard potential index are obtained through principal component analysis; the exposure potential index and hazard potential index are processed to obtain the priority control index; based on the priority control index, sorting is carried out to screen OPE compounds that meet the preset conditions, and the screening results are determined as the priority control targets. This application can simultaneously identify key pollutants with high exposure risks and high hazard risks, reveal the main risk drivers of different OPEs, and effectively support the screening and priority control decision-making of emerging pollutants.

Description

Technical Field

[0001] This application relates to the field of environmental monitoring and data analysis technology, and in particular to a method for prioritizing the screening of OPEs based on surface water concentration data. Background Technology

[0002] Organophosphate flame retardants (OPEs) are a class of additives widely used in plastics, textiles and electronic products. With their large-scale production and use, they have been widely detected in surface water environments and have shown high exposure levels, becoming one of the important emerging pollutants.

[0003] Existing studies are mostly based on monitoring data from specific watersheds, urban river networks, or local sewage outlets. However, these screening methods have significant limitations. First, their regional representativeness is relatively limited. Due to differences in the production layout, usage type, and emission pathways of OPEs in different regions, priority control lists derived from local data are often difficult to generalize to a national scale, making it difficult to unify and coordinate management measures. In addition, screening results are easily affected by regional bias. When data comes from only a few regions or specific types of water bodies, the risks of some OPEs may be overestimated or underestimated, thus affecting the rational allocation of pollutant control resources.

[0004] Meanwhile, the prioritization methods in existing evaluation systems also have a certain degree of subjectivity. Specifically, some indicators, such as weight setting and judgment of exposure importance, often rely on researchers' experience or expert opinions. This not only allows for greater room for human intervention but also results in weak repeatability and comparability of results, ultimately hindering the formation of objective and unified priority control standards. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, this application provides a priority control OPEs screening method based on surface water concentration data, which solves the technical problem that the evaluation results are highly subjective and have poor comparability due to the reliance on expert experience to assign weights to indicators, making it difficult to form a unified priority control standard.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the main technical solutions adopted in this application include:

[0009] In one aspect, embodiments of this application provide a method for prioritizing the screening of OPEs based on surface water concentration data.

[0010] The priority control OPEs screening method based on surface water concentration data proposed in this application mainly includes:

[0011] S100. Based on the given organic phosphate flame retardant (OPE) exposure database, obtain monitoring data for each OPE compound within a specified time period; wherein, the monitoring data includes: the name, CAS number, environmental concentration, and detection frequency of each OPE compound;

[0012] S200. Based on the environmental concentration and detection frequency in the monitoring data, the exposure potential index is obtained through principal component analysis.

[0013] S300. For each OPE compound, based on the monitoring data, obtain quantitative data for each OPE compound in multiple dimensions based on environmental behavior characteristics and ecotoxicity characteristics, and perform principal component analysis on the quantitative data to obtain the hazard potential index.

[0014] S400. For each OPE compound, process the exposure potential index and hazard potential index of that OPE compound to obtain the priority control index for each OPE compound.

[0015] S500: Sort the compounds based on the priority control index, screen out OPEs compounds that meet the preset conditions, and determine the screening results as priority control objects.

[0016] Specifically, the exposure potential index, obtained through principal component analysis based on the environmental concentration and detection frequency in the monitoring data, includes:

[0017] The environmental concentrations were normalized, and the environmental concentrations and detection frequencies of each OPE compound were converted into relative proportions based on the maximum and minimum values, respectively.

[0018] Based on the relative ratio of the environmental concentration to the detection frequency, principal component analysis is used to determine the respective weights of the environmental concentration and the detection frequency, and the exposure potential index of each OPE compound is generated based on the weights.

[0019] Optionally, performing principal component analysis on the quantified data to obtain the hazard potential index includes:

[0020] The quantitative data based on multiple dimensions of environmental behavior characteristics and ecotoxicity characteristics are normalized. Based on the normalized data, principal component analysis is used to determine the weights of each of the multiple dimensions based on environmental behavior characteristics and ecotoxicity characteristics, and the hazard potential index of each OPE compound is generated according to the weights.

[0021] Optionally, the acquisition of quantitative data for each OPE compound in three dimensions—environmental persistence, bioaccumulation, and toxic effects—includes:

[0022] Multiple dimensions of environmental behavior and ecotoxicity characteristics, including: environmental persistence, bioaccumulation, and toxic effects;

[0023] Based on the molecular structure of each OPE compound, a biodegradability prediction model is used to evaluate and predict the final biodegradation time of each OPE compound, serving as quantitative data on environmental durability.

[0024] Based on the molecular structure of each OPE compound, the logarithm of the octanol / water partition coefficient of each OPE compound is obtained as quantitative data on bioaccumulation.

[0025] Acute toxicity data of algae and invertebrates were obtained separately. The missing acute toxicity data were predicted using a quantitative structure-activity relationship model. Endocrine disruption effect data of vertebrates were obtained through molecular docking simulation. The acute toxicity data and endocrine disruption effect data were combined to obtain quantitative data of toxicity effects.

[0026] Optionally, the step of using a quantitative structure-activity relationship model to predict missing acute toxicity data, and the step of obtaining endocrine disruption effect data in vertebrates through molecular docking simulation calculations, include:

[0027] For algae, based on the molecular structure of each OPE compound, the acute toxicity data of each OPE compound to the algae are predicted using a pre-constructed quantitative structure-activity relationship model for algae; the quantitative structure-activity relationship model for algae is constructed based on a semi-empirical quantum chemical descriptor.

[0028] For invertebrates, based on the molecular structure of each OPE compound, the acute toxicity data of each OPE compound to the invertebrates are predicted using a quantitative structure-activity relationship model.

[0029] For vertebrates, the three-dimensional structures of selected vertebrate hormone receptor proteins are obtained from a protein structure database. Molecular docking software is used to simulate the docking of the molecular structure of each OPE compound with the receptor protein, and the docking score, which characterizes the binding strength, is calculated as the endocrine interference effect data.

[0030] Optionally, the pre-constructed quantitative structure-activity relationship (QSPR) model for algae is a linear prediction equation established using a stepwise multiple linear regression method based on quantum chemical descriptors and acute toxicity data of OPEs compounds in a publicly available database. The training process of the QSPR model for algae includes:

[0031] A1. Using pre-defined algae as model organisms, and following pre-defined acute toxicity experimental conditions, quantum chemical descriptors of OPEs compounds and corresponding acute toxicity data are screened and collected from public databases to form a training dataset.

[0032] A2. Randomly divide the training dataset into a training set and a validation set according to a preset ratio;

[0033] A3. Based on the training set, a linear prediction equation between the quantum chemical descriptor and the acute toxicity data is established using the stepwise multiple linear regression method.

[0034] A4. Based on the validation set, the goodness of fit and predictive ability of the linear prediction equation are validated using the coefficient of determination and external validation parameters.

[0035] Optionally, the step of using molecular docking software to dock the molecular structure of each OPE compound with the receptor protein and calculating the docking score characterizing the binding strength as the endocrine disruption effect data includes:

[0036] Using zebrafish estrogen receptors and androgen receptors as target proteins, the binding score of each OPE compound to the target protein was calculated using molecular docking software, and this score was used as data for the endocrine disruption effect.

[0037] Specifically, for each OPE compound, the exposure potential index and hazard potential index of that OPE compound are processed to obtain the priority control index for each OPE compound, including:

[0038] For each OPE compound, the exposure potential index and hazard potential index of that OPE compound are normalized separately. The two normalized indices are then multiplied to obtain the priority control index for each OPE compound.

[0039] Specifically, the OPEs compounds include halogenated, aromatic, and alkyl organophosphate flame retardants.

[0040] Secondly, embodiments of this application provide a computing device for a priority control OPEs screening method based on surface water concentration data.

[0041] This application provides a computing device comprising a memory, a processor, and a computer program stored in the memory, characterized in that the processor executes the computer program to implement any of the priority control OPEs screening methods based on surface water concentration data according to claims 1 to 9.

[0042] (III) Beneficial Effects

[0043] The beneficial effects of this application are as follows: The priority control OPEs screening method based on surface water concentration data in this application, by acquiring monitoring data from a pre-established exposure database and calculating the exposure potential index and hazard potential index based on principal component analysis, and then processing the two to obtain the priority control index and sorting and screening, can achieve an objective and comprehensive quantitative evaluation of the exposure risk and hazard risk of OPEs of various structural types, compared with evaluation methods that rely on expert experience for subjective weighting or fail to integrate the two dimensions of exposure and hazard. It avoids the subjective bias of manually setting weights, and achieves the effect of automatically identifying priority control pollutants and revealing the main risk drivers of different OPEs, thereby providing core technical support for establishing a nationally unified, scientific and reliable priority control list. Attached Figure Description

[0044] Figure 1 This is a schematic diagram illustrating the steps of a priority control OPEs screening method based on surface water concentration data according to an embodiment of this application;

[0045] Figure 2 This is a flowchart illustrating a method for prioritizing OPEs screening based on surface water concentration data according to an embodiment of this application. Detailed Implementation

[0046] To better explain and facilitate understanding of this application, the following detailed description of the application is provided in conjunction with the accompanying drawings and specific embodiments.

[0047] The proposed method for screening priority control organophosphate flame retardants based on surface water concentration data addresses the core technical problem in related technologies: poor comparability and narrow applicability of priority ranking results due to regional data limitations and subjective weighting. This method integrates exposure monitoring data of OPE (Organophosphate Esters) compound concentrations in surface water, combining hazard attributes such as environmental persistence, bioaccumulation, and multi-species toxicity effects. Principal component analysis is used to objectively weight each indicator, and risk ranking is achieved by calculating the product of the exposure potential index and the hazard potential index. This method avoids the subjectivity of weight setting, improves the robustness of the assessment results and its applicability at the national scale, and can systematically identify priority control pollutants with both high exposure levels and high hazard potential, providing a unified and scientific decision-making basis for environmental supervision and risk management of organophosphate flame retardants in surface water.

[0048] To better understand the above technical solutions, exemplary embodiments of this application will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of this application are shown in the drawings, it should be understood that this application can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application can be understood more clearly and thoroughly, and that the scope of this application can be fully conveyed to those skilled in the art.

[0049] Example 1

[0050] Figure 1 This is a schematic diagram illustrating the steps of a priority control OPEs screening method based on surface water concentration data according to some embodiments of this application, such as... Figure 1 As shown, the priority control OPEs screening method based on surface water concentration data in this embodiment includes:

[0051] S100. Based on the given OPEs exposure database, obtain monitoring data for each OPE compound within a specified time period; wherein, the monitoring data includes: the name, CAS number, environmental concentration and detection frequency of each OPE compound.

[0052] S200: Based on the environmental concentration and detection frequency in the monitoring data, the exposure potential index is obtained through principal component analysis.

[0053] S300. For each OPE compound, based on monitoring data, quantitative data of each OPE compound in three dimensions of environmental persistence, bioaccumulation and toxic effects are obtained. Principal component analysis is performed on the quantitative data to obtain the hazard potential index.

[0054] S400. For each OPE compound, process the exposure potential index and hazard potential index of that OPE compound to obtain the priority control index for each OPE compound.

[0055] S500: Sort by priority control index, screen out OPEs compounds that meet preset conditions, and determine the screening results as priority control targets.

[0056] This embodiment integrates the exposure potential index and the hazard potential index, and employs principal component analysis for objective weighting, effectively avoiding the subjectivity and regional bias of manually setting weights in traditional methods. This method can comprehensively reflect the exposure level and potential ecological hazards of compounds in actual aquatic environments, supporting unified and comparable risk ranking of dozens to hundreds of OPEs, significantly improving the scientific rigor and applicability of screening results, and providing a reliable technical basis for the control decisions and standard setting of organophosphate flame retardants in my country's surface water.

[0057] Example 2

[0058] like Figure 2 As shown in this embodiment, the priority control OPEs screening method based on surface water concentration data includes:

[0059] S100. Based on the given OPEs exposure database, obtain monitoring data for each OPE compound within a specified time period; wherein, the monitoring data includes: the name, CAS number, environmental concentration and detection frequency of each OPE compound.

[0060] S200: Based on the environmental concentration and detection frequency in the monitoring data, the exposure potential index is obtained through principal component analysis.

[0061] S300. For each OPE compound, based on monitoring data, quantitative data of each OPE compound in three dimensions of environmental persistence, bioaccumulation and toxic effects are obtained. Principal component analysis is performed on the quantitative data to obtain the hazard potential index.

[0062] S400. For each OPE compound, process the exposure potential index and hazard potential index of that OPE compound to obtain the priority control index for each OPE compound.

[0063] S500: Sort by priority control index, screen out OPEs compounds that meet preset conditions, and determine the screening results as priority control targets.

[0064] Monitoring data on OPEs in Chinese surface water from 2020 to 2025 were collected from publicly available literature databases. The names, CAS numbers (Chemical Abstracts Service Registry Numbers), detection frequencies, and environmental concentrations of each compound were extracted to establish a nationwide OPEs exposure database. Compounds lacking sufficient monitoring records were removed to ensure data stability.

[0065] Using environmental concentration and detection frequency as exposure indicators, the concentration data were statistically standardized, and principal component analysis was employed to perform dimensionality reduction analysis on the indicators, extracting principal characteristic variables to obtain the exposure potential index of each OPE. This process automatically determines the indicator weights through a statistical model, eliminating the need for subjective assignment and thus avoiding human bias.

[0066] The hazard potential of compounds is characterized by three dimensions: persistence (P), bioaccumulation (B), and toxicity (T). The three types of indicators (PBT) are standardized and integrated to calculate the hazard potential index, and the principal component analysis method is also used.

[0067] The exposure potential index and the hazard potential index are normalized separately, and the product of the two is used to calculate the priority control index. The results are then ranked according to their values, with the index ranking higher being more noteworthy.

[0068] The priority control index based on multi-parameter integration constructed in this embodiment enables a comprehensive assessment of the exposure risk and hazard risk of OPEs. Principal component analysis is used to objectively assign weights to five indicators: detection frequency, environmental concentration, environmental persistence, bioaccumulation, and toxicity. This avoids subjective biases that may be introduced by manually setting weights, ensuring the stability and consistency of the evaluation results.

[0069] The results show that different indicators contribute differently to pollutant priority, with detection frequency contributing the most, followed by environmental persistence, environmental concentration, and biotoxicity, while bioaccumulation contributes less. This indicates that the prevalence of pollutants in water bodies and their recalcitrant nature in the environment have a more significant impact on risk ranking. The index system established in this embodiment can reflect the main risk drivers of different OPEs.

[0070] This embodiment uses the detection frequency and concentration of various organophosphate flame retardants (OPEs) in water as exposure index parameters, and calculates persistence data, bioaccumulation data, and ecotoxicity data as hazard index parameters. Utilizing multi-dimensional parameters, a comprehensive risk priority assessment of various OPEs in water is conducted, ranking the risk intensity of different OPEs. This approach has advantages such as objective weighting, highly visualized results, and broad system adaptability. It can yield risk ranking results that more closely reflect real-world aquatic ecosystem exposure scenarios, based on a comprehensive consideration of exposure levels, environmental persistence, and various biotoxic effects.

[0071] Specifically, based on the environmental concentration and detection frequency in the monitoring data, the exposure potential index is obtained through principal component analysis, including:

[0072] Environmental concentrations were normalized, and the environmental concentration and detection frequency of each OPE compound were converted into relative proportions based on their maximum and minimum values. Based on these relative proportions, principal component analysis was used to determine the weights of each environmental concentration and detection frequency. An exposure potential index for each OPE compound was then generated based on these weights.

[0073] Experimental testing was conducted on surface water in the assessed area. The detection frequency and concentration of each organophosphate flame retardant were used as the exposure risk factor for that respective flame retardant. The environmental concentrations of each OPE compound were normalized using the following formula:

[0074] ,

[0075] Where, x normalized x represents the normalized value. min Let x represent the minimum value. max This represents the maximum value.

[0076] Principal component analysis was used to process the normalized environmental concentration and detection frequency. The eigenvectors corresponding to two variables in the first principal component were selected as their weight coefficients w. c and w DF .

[0077] The Exposure Potential Index (EI) is calculated using the following formula:

[0078] ,

[0079] Among them, w c The weighting coefficient representing the environmental concentration, w DF The weighting coefficient represents the detection frequency, Concentration represents the normalized value of the concentration, and Detection Frequency represents the detection frequency, which itself does not need to be normalized in the range of 0-1.

[0080] This embodiment normalizes environmental concentration and detection frequency and automatically determines their weights in the exposure potential index using principal component analysis. This method eliminates the uncertainty of relying on expert experience or subjective assignment in traditional assessments, ensuring the objectivity and statistical robustness of the exposure evaluation process. This more accurately reflects the actual exposure distribution characteristics of various OPEs across the country, enhancing the reliability of screening results and the comparability between regions.

[0081] Specifically, principal component analysis is performed on the quantitative data to obtain a hazard potential index, including:

[0082] The quantitative data based on multiple dimensions of environmental behavior characteristics and ecotoxicity characteristics are normalized. Based on the normalized data, principal component analysis is used to determine the weights of each of the multiple dimensions based on environmental behavior characteristics and ecotoxicity characteristics, and the hazard potential index of each OPE compound is generated according to the weights.

[0083] Persistence, bioaccumulation, and ecotoxicity data were calculated as hazard risk factors for various organophosphate flame retardants. Data from multiple dimensions based on environmental behavior and ecotoxicity characteristics were quantified, and the three quantified data were normalized using a normalization formula. Principal component analysis was used to process the normalized environmental persistence, bioaccumulation, and toxicity effect data. The eigenvectors corresponding to the three variables in the first principal component were selected as the weight coefficients w for environmental persistence, bioaccumulation, and toxicity effects, respectively. P w B and w T .

[0084] The hazard potential index (HI) is calculated using the following formula:

[0085] ,

[0086] Among them, w P w B and w T These are the weighting coefficients for environmental persistence, bioaccumulation, and toxicity, respectively. Persistence represents the normalized value of environmental persistence, Bioaccumulation represents the normalized value of bioaccumulation, and Toxicity represents the normalized value of toxicity.

[0087] This embodiment normalizes and assigns weights to three types of indicators—persistence, bioaccumulation, and toxicity—through principal component analysis. This method achieves systematic integration and objective quantification of multi-dimensional hazard attributes, avoids evaluation bias caused by inconsistent indicator weight settings, improves the overall and scientific nature of hazard assessment, and enables fair comparison of OPEs of different structural types under the same standard, providing a more comprehensive basis for risk management.

[0088] Optionally, quantitative data for each OPE compound in three dimensions—environmental persistence, bioaccumulation, and toxic effects—are obtained, including: based on the molecular structure of each OPE compound, an evaluation using a biodegradability prediction model is performed to predict the final biodegradation time of each OPE compound, which serves as quantitative data for environmental persistence.

[0089] Based on the molecular structure of each OPE compound, the logarithm of the octanol / water partition coefficient of each OPE compound is obtained as quantitative data on bioaccumulation.

[0090] Acute toxicity data of algae and invertebrates were obtained separately. The missing acute toxicity data were predicted using a quantitative structure-activity relationship model. Endocrine disruption effect data of vertebrates were obtained through molecular docking simulation. The acute toxicity data and endocrine disruption effect data were combined to obtain quantitative data of toxicity effects.

[0091] Environmental durability: Environmental durability was evaluated based on the biodegradability model (BIOWIN3). The BIOWIN3 parameters obtained from the BIOWIN module in EPISuite v4.1 were used to assess the final biodegradation time of different OPEs compounds as environmental durability data.

[0092] Bioaccumulation: The bioaccumulation potential of aquatic organisms was characterized by the octanol / water partition coefficient (log kow); the log kow was obtained through the BCFBAF v3.01 module in EPISuite to assess the absorption, distribution, metabolism, excretion and toxicity of organophosphate flame retardants as bioaccumulation data.

[0093] Toxicity effects: The acute toxicity of algae and invertebrates, as well as the endocrine disruption effects of vertebrates, were considered. When experimental toxicity data were insufficient, a QSAR (Quantitative Structure-Activity Relationship) model was used for prediction and completion. Endocrine effects were characterized by binding affinity to receptor proteins using biomolecular docking technology. The algae *Scenedesmus subspicatus*, the invertebrate *Daphniamagna*, and the vertebrate *Danio rerio* were selected as representative species for toxicity evaluation of these three typical organisms.

[0094] The acute toxicity data of organophosphate flame retardants to algae and invertebrates were calculated using an ecological structure-activity relationship prediction model, while the endocrine disruption effect on vertebrates was taken into account.

[0095] This embodiment combines a biodegradability prediction model, octanol / water partition coefficient, quantitative structure-activity relationship model, and molecular docking technology to systematically and efficiently obtain quantitative data on OPEs in three key dimensions: environmental persistence, bioaccumulation, and toxic effects. It is particularly suitable for compounds with missing experimental data, achieving comprehensive coverage and scientific prediction of the hazardous properties of OPEs, and enhancing the feasibility and data integrity of the method in practical applications.

[0096] Optionally, quantitative structure-activity relationship models are used to predict missing acute toxicity data, and endocrine disruption effect data in vertebrates are obtained through molecular docking simulations, including:

[0097] For algae, based on the molecular structure of each OPE compound, a pre-constructed quantitative structure-activity relationship (QSPR) model for algae is used to predict the acute toxicity data of each OPE compound to algae; the QSPR model for algae is constructed based on a semi-empirical quantum chemical descriptor.

[0098] For invertebrates, based on the molecular structure of each OPE compound, the acute toxicity data of each OPE compound to invertebrates were predicted using a quantitative structure-activity relationship model. The representative invertebrate was Daphnia magna, and the acute toxicity data of each OPE compound to Daphnia magna were calculated using the Daphnia magna 48hLC50 model in the TEST software.

[0099] For vertebrates, the three-dimensional structures of selected vertebrate hormone receptor proteins are obtained from protein structure databases. Molecular docking software is used to simulate the docking of the molecular structure of each OPE compound with the receptor protein, and the docking score, which characterizes the binding strength, is calculated as data on endocrine interference effects.

[0100] The three-dimensional structures of biological receptor proteins were obtained from the AlphaFold Protein Structure Database (a public database for protein structure prediction). The Dock-Ligands (Libdock, molecular docking module) tool of Discovery Studio 2020 software (molecular simulation software) was used to dock the additives with the receptor proteins, calculate the molecular binding energy, and use the libdock score as the calculation parameter. The libdock score is a scoring value unique to the molecular docking module. The higher the score, the stronger the binding affinity.

[0101] Algae, invertebrates, and vertebrates are considered equally important; therefore, Toxicity's calculation formula is:

[0102] ,

[0103] Wherein, Dr toxicity is the docking score with both hormone receptors, calculated using the following formula:

[0104] ,

[0105] In the two formulas above, Ss toxicity represents the normalized acute toxicity data of *Scenedesmus subspicatus*; Dm toxicity represents the normalized acute toxicity data of *Daphnia magna*; D Score represents the docking score, and DScore represents the docking score. max The D score represents the maximum docking score between the two hormone receptors. min This represents the minimum docking score between the two hormone receptors.

[0106] This embodiment employs quantitative structure-activity relationship models and molecular docking technology to predict toxicity in algae, invertebrates, and vertebrates, respectively. This not only compensates for the lack of experimental data but also allows for a more accurate assessment of the potential impact of OPEs on organisms at different trophic levels. In particular, by simulating the binding ability with zebrafish hormone receptors to evaluate endocrine disruption effects, the toxicity assessment is enhanced in terms of its hierarchical and biological relevance, making the hazard index more ecologically representative.

[0107] Optionally, the pre-constructed quantitative structure-activity relationship (QSPR) model for algae is a linear prediction equation established using a stepwise multiple linear regression method based on the quantum chemical descriptors and acute toxicity data of OPEs compounds in a publicly available database. The training process of the QSPR model for algae includes:

[0108] A1. Using pre-defined algae as model organisms, and following pre-defined acute toxicity experimental conditions, quantum chemical descriptors of OPEs compounds and corresponding acute toxicity data are screened and collected from public databases to form a training dataset.

[0109] A2. Randomly divide the training dataset into a training set and a validation set according to a preset ratio.

[0110] A3. Based on the training set, a linear prediction equation between quantum chemical descriptors and acute toxicity data is established using the stepwise multiple linear regression method.

[0111] A4. Based on the validation set, the coefficient of determination and external validation parameters are used to verify the goodness of fit and predictive ability of the linear prediction equation.

[0112] The model organism selected for the algal QSAR model is *Scenedesmus obliquus*. Acute toxicity data of *Scenedesmus obliquus* were collected from existing databases. The selection criteria for acute toxicity data were: freshwater experimental conditions, toxicity effect measured by population growth rate, experimental observation period of 3 days, and toxicity endpoint measured by EC50. 50 (Half Maximal Effective Concentration) If multiple toxicity data exist for the same substance, their geometric mean is taken as the final acute toxicity data.

[0113] Due to limited data on algae experiments, acute toxicity data of 12 OPEs to Scenedesmus obliquus were collected and randomly divided into training and validation sets at a ratio of 5:1.

[0114] Quantum chemical descriptors of compounds were collected. These descriptors can be directly derived from molecular wave functions and applied to quantitative structure-activity relationships, providing a precise quantitative description of molecular structure and its chemical properties. After obtaining molecular structure files from the Pubchem database, molecular energy optimization was performed using the Gaussian interface method in Chem Bio3D software. Then, the semi-empirical molecular descriptors were calculated using the PM7 method in MOPAC (2016). The names, abbreviations, and units of the 13 quantum chemical descriptors are shown in Table 1.

[0115] Then, a QSAR model for the descriptors and toxicity data was constructed using stepwise multiple linear regression (MLR). The fitted QSAR model is as follows:

[0116] ,

[0117] The model selects the total molecular energy and the net charge of the most negative oxygen atom as descriptors. The total molecular energy refers to the overall molecular energy calculated using semi-empirical quantum chemistry methods. Its value includes major components such as electron energy, core repulsion energy, and nuclear repulsion energy, and is measured in electron volts (eV). Total energy is an indicator of the overall stability of the molecule; the lower the energy, the more stable the molecular structure. − The net charge of the oxygen atom with the most negative net charge in a molecule reflects its nucleophilicity, hydrogen bond acceptor ability, and tendency to interact with targets (such as receptors and enzymes). TE indicates that the toxicity of OPEs to algae may be related to molecular polarity, which affects the absorption of compounds by organisms and their distribution in the environment. The most negative oxygen atom in OPEs is always a double-bonded O atom on its phosphate functional group. Phosphate groups are an important characteristic structure of OPEs, suggesting that the toxicity of OPEs should be related to their phosphate groups.

[0118]

[0119] Table 1. 13 quantum chemical descriptors

[0120] Total molecular energy (OPE): Calculated using semi-empirical quantum chemical methods, it characterizes the overall stability of the molecule; a lower value generally indicates a more stable molecular structure. This energy is expressed in electron volts (eV). This parameter suggests that the toxicity of OPEs to algae may be related to molecular polarity, which affects the absorption of compounds by organisms and their distribution in the environment.

[0121] Net charge of the most negative oxygen atom: This refers to the charge value of the oxygen atom with the largest negative net charge among all oxygen atoms in the molecule. In OPEs, this atom is always a double-bonded oxygen atom on its phosphate functional group. This charge effectively reflects the nucleophilicity, hydrogen bond acceptor ability, and interaction tendency of this oxygen atom with biological targets. Phosphate groups are a characteristic structure of OPEs, and this parameter reveals that the toxic effects of OPEs are closely related to their phosphate groups.

[0122] The model goodness of fit was evaluated using the coefficient of determination (R²) and root mean square error (RMSE), the robustness of the model was examined using the bootstrap correlation coefficient (QBOOT²), and external validation parameters were used. The predictive ability of the model was evaluated using the R² method. The predictive model showed good statistical performance and good data fit (R² > 0.6), and can be used to supplement toxicity data. All variable inflation factors (VIF) were less than 10, and there was no collinearity problem.

[0123] This embodiment constructs a quantitative structure-activity relationship model for algae based on semi-empirical quantum chemical descriptors and employs a combination of stepwise multiple linear regression and external validation to ensure that the model has good fit and predictive ability. It can reliably supplement algal toxicity data, enhance the data foundation and scientific rigor of the entire toxicity assessment system, and provide more solid model support for the ecological risk assessment of OPEs.

[0124] Optionally, molecular docking software is used to dock the molecular structure of each OPE compound with the receptor protein, and the docking score characterizing the binding strength is calculated as endocrine disruption effect data. This includes: using zebrafish estrogen receptor and androgen receptor as target proteins, the binding score between each OPE compound and the target protein is simulated and calculated using molecular docking software as endocrine disruption effect data.

[0125] The zebrafish estrogen receptor ER (UniProt accession number P57717, and the predicted structural model AF-P57717-F1 provided by the AlphaFold database) and the zebrafish androgen receptor AR (UniProt accession number A4GT83, structural model: AlphaFold AF-A4GT83-F1) were selected.

[0126] This embodiment directly simulates the endocrine disruption that OPEs may cause, thereby more accurately reflecting their potential harm to aquatic vertebrates. By calculating the docking score to quantify the binding strength, it achieves an objective and quantitative evaluation of the endocrine disruption effect of the compound, making up for the shortcomings of traditional toxicity tests in assessing such effects.

[0127] Specifically, for each OPE compound, its exposure potential index and hazard potential index are processed to obtain the priority control index for each OPE compound, including:

[0128] For each OPE compound, the exposure potential index and hazard potential index of that OPE compound are normalized separately. The two normalized indices are then multiplied to obtain the priority control index for each OPE compound.

[0129] Principal component analysis (PCA) was used to analyze two exposure parameters, with the normalized score defined as the Exposure Indicator (EI) to describe the exposure potential of pollutants. PCA was also used to analyze three hazard parameters, extracting the first two principal components as comprehensive weighting factors. A formula for calculating the comprehensive hazard index was constructed based on their eigenvector loading values. The risk ranking of different organophosphate flame retardants in water bodies was derived based on the risk data.

[0130] After calculating the Exposure Potential Index (EI) and Hazard Potential Index (HI) for each OPE compound, to eliminate the dimensional differences between the two indices and ensure the comparability of the multiplication operations, a minimum-maximum normalization method consistent with the aforementioned parameter normalization was used to normalize both EI and HI. Specifically, all OPE compounds to be evaluated were examined to determine the minimum value of EI (EI0). min ) and maximum value (EI) max ), and the minimum value of HI (HI) min ) and maximum value (HI) max Using the above normalization formula, after normalizing the exposure potential index and hazard potential index of each compound, the priority control index (PI) is calculated using the following formula:

[0131] ,

[0132] The OPEs were sorted by their Principal Indices (PIs). Based on the ranking results, the OPEs were divided into a High Concern Group and a General Concern Group. Ten priority control substances were selected from the 27 OPEs. These included various structural types such as halogenated, aromatic, and alkyl compounds, demonstrating the universality and applicability of this embodiment under conditions of structural type differences. For example, V6 (2,2-bis(chloromethyl)trimethylene bis(bis(2-chloroethyl)phosphate), also commonly known as flame retardant V6, CAS No.: 38051-10-4), although with relatively weak bioaccumulation, ranks highest in priority due to its high toxicity, high persistence, and significant environmental exposure; while RDP (Resorcinol bis(diphenyl phosphate), CAS No.: 57583-54-7) and BDP (Bisphenol A bis(diphenyl phosphate), CAS No.: 181028-79-5), although detected less frequently, are still included in the priority control range due to their high environmental concentration and hazard potential; TIBP (Triisobutyl Triisobutyl phosphate (CAS No.: 126-71-6) exhibits low hazard but high exposure levels, suggesting it should be a key monitoring target.

[0133] This embodiment achieves an effective integration of exposure and hazard attributes using a simple mathematical model, which avoids the uncertainty caused by complex calculations and can intuitively reflect the overall risk level of the compound, making the ranking results clear and easy to understand.

[0134] Specifically, various OPEs compounds, including halogenated, aromatic, and alkyl organophosphate flame retardants.

[0135] Organophosphate flame retardants (OPEs) encompass their main structural types, including halogenated OPEs, aromatic OPEs, and alkyl OPEs. These three classes of compounds differ significantly in molecular structure, physicochemical properties, and environmental behavior. For example, halogenated OPEs typically exhibit high environmental persistence and potential toxicity; aromatic OPEs may display varying adsorption and bioaccumulation characteristics; and alkyl OPEs often possess high water solubility and migration capacity. Traditional evaluation methods based on localized data or subjective weighting are prone to evaluation bias due to the specific properties of a particular type of compound.

[0136] This embodiment is applicable to organophosphate flame retardants with various structural types such as halogenated, aromatic, and alkyl, and has strong universality and structural inclusiveness. It can comprehensively cover common OPE pollutants in current surface water, ensuring that compounds with different chemical properties can be fairly ranked in the same evaluation system, thereby supporting the development of a more comprehensive and systematic priority control list.

[0137] Example 3

[0138] This application also provides a computing device, including a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement any of the priority control OPEs screening methods based on surface water concentration data in embodiments 1 and 2 above.

[0139] This application also proposes a readable storage medium storing a program or instructions that, when executed by a processor, implement the steps of any of the above-described methods for prioritizing OPEs screening based on surface water concentration data.

[0140] Among them, readable storage media include read-only memory (ROM), random access memory (RAM), magnetic disks or optical disks, etc.

[0141] The readable storage medium provided in this application embodiment implements the steps of any of the above-described methods for prioritizing OPE screening based on surface water concentration data when the program or instructions are executed by a processor. Therefore, the readable storage medium includes all the beneficial effects of the above-described methods for prioritizing OPE screening based on surface water concentration data.

[0142] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0143] In the description of this application, it should be understood that the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0144] In this application, unless otherwise expressly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0145] In this application, unless otherwise expressly specified and limited, "above" or "below" the second feature can mean that the first and second features are in direct contact, or that they are in indirect contact through an intermediate medium. Furthermore, "above," "on top of," and "over" the second feature can mean that the first feature is directly above or diagonally above the second feature, or simply that the first feature is at a higher horizontal level than the second feature. "Below," "below," and "under" the second feature can mean that the first feature is directly below or diagonally below the second feature, or simply that the first feature is at a lower horizontal level than the second feature.

[0146] In the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0147] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make modifications, alterations, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for prioritizing the selection of OPEs based on surface water concentration data, characterized in that, include: Based on the given OPEs exposure database, monitoring data for each OPE compound was obtained; wherein, the monitoring data included: the name, CAS number, environmental concentration and detection frequency of each OPE compound; Based on the environmental concentration and detection frequency in the monitoring data, the exposure potential index is obtained through principal component analysis. For each OPE compound, based on the monitoring data, quantitative data of each OPE compound based on multiple dimensions of environmental behavior characteristics and ecotoxicity characteristics are obtained. Principal component analysis is performed on the quantitative data to obtain the hazard potential index. For each OPE compound, the exposure potential index and hazard potential index of that OPE compound are processed to obtain the priority control index for each OPE compound. Based on the priority control index, OPEs compounds that meet the preset conditions are sorted and selected, and the selection results are determined as priority control targets.

2. The method according to claim 1, characterized in that, The exposure potential index, obtained through principal component analysis based on the environmental concentration and detection frequency in the monitoring data, includes: The environmental concentrations were normalized, and the environmental concentrations and detection frequencies of each OPE compound were converted into relative proportions based on the maximum and minimum values, respectively. Based on the relative ratio of the environmental concentration to the detection frequency, principal component analysis is used to determine the respective weights of the environmental concentration and the detection frequency, and the exposure potential index of each OPE compound is generated based on the weights.

3. The method according to claim 1, characterized in that, The principal component analysis of the quantified data to obtain the hazard potential index includes: The quantitative data based on multiple dimensions of environmental behavior characteristics and ecotoxicity characteristics are normalized. Based on the normalized data, principal component analysis is used to determine the weights of each of the multiple dimensions based on environmental behavior characteristics and ecotoxicity characteristics, and the hazard potential index of each OPE compound is generated according to the weights.

4. The method according to claim 1, characterized in that, The acquisition of quantitative data for each OPE compound based on multiple dimensions of environmental behavior and ecotoxicity characteristics includes: Multiple dimensions of environmental behavior and ecotoxicity characteristics, including: environmental persistence, bioaccumulation, and toxic effects; Based on the molecular structure of each OPE compound, a biodegradability prediction model is used to evaluate and predict the final biodegradation time of each OPE compound, serving as quantitative data on environmental durability. Based on the molecular structure of each OPE compound, the logarithm of the octanol / water partition coefficient of each OPE compound is obtained as quantitative data on bioaccumulation. Acute toxicity data of algae and invertebrates were obtained separately. The missing acute toxicity data were predicted using a quantitative structure-activity relationship model. Endocrine disruption effect data of vertebrates were obtained through molecular docking simulation. The acute toxicity data and endocrine disruption effect data were combined to obtain quantitative data of toxicity effects.

5. The method according to claim 4, characterized in that, The method of predicting missing acute toxicity data using a quantitative structure-activity relationship model, and the method of obtaining endocrine disruption effect data in vertebrates through molecular docking simulation calculations, include: For algae, based on the molecular structure of each OPE compound, the acute toxicity data of each OPE compound to the algae are predicted using a pre-constructed quantitative structure-activity relationship model for algae; the quantitative structure-activity relationship model for algae is constructed based on a semi-empirical quantum chemical descriptor. For invertebrates, based on the molecular structure of each OPE compound, the acute toxicity data of each OPE compound to the invertebrates are predicted using a quantitative structure-activity relationship model. For vertebrates, the three-dimensional structures of selected vertebrate hormone receptor proteins are obtained from a protein structure database. Molecular docking software is used to simulate the docking of the molecular structure of each OPE compound with the receptor protein, and the docking score, which characterizes the binding strength, is calculated as the endocrine interference effect data.

6. The method according to claim 5, characterized in that, The pre-constructed quantitative structure-activity relationship (QSPR) model for algae is a linear prediction equation established using a stepwise multiple linear regression method based on quantum chemical descriptors and acute toxicity data of OPEs compounds in a publicly available database. The training process of the QSPR model for algae includes: A1. Using pre-defined algae as model organisms, and following pre-defined acute toxicity experimental conditions, quantum chemical descriptors of OPEs compounds and corresponding acute toxicity data are screened and collected from public databases to form a training dataset. A2. Randomly divide the training dataset into a training set and a validation set according to a preset ratio; A3. Based on the training set, a linear prediction equation between the quantum chemical descriptor and the acute toxicity data is established using the stepwise multiple linear regression method. A4. Based on the validation set, the goodness of fit and predictive ability of the linear prediction equation are validated using the coefficient of determination and external validation parameters.

7. The method according to claim 5, characterized in that, The process involves using molecular docking software to dock the molecular structure of each OPE compound with the receptor protein, calculating a docking score characterizing the binding strength, and using this score as data on the endocrine disruption effect. This includes: Using zebrafish estrogen receptors and androgen receptors as target proteins, the binding score of each OPE compound to the target protein was calculated using molecular docking software, and this score was used as data for the endocrine disruption effect.

8. The method according to claim 1, characterized in that, For each OPE compound, the exposure potential index and hazard potential index of that OPE compound are processed to obtain the priority control index for each OPE compound, including: For each OPE compound, the exposure potential index and hazard potential index of that OPE compound are normalized separately. The two normalized indices are then multiplied to obtain the priority control index for each OPE compound.

9. The method according to claim 1, characterized in that, The OPEs compounds include halogenated, aromatic, and alkyl organophosphate flame retardants.

10. A computing device, comprising a memory, a processor, and a computer program stored in the memory, characterized in that, The processor executes the computer program to implement any of the preferred OPEs screening methods based on surface water concentration data according to claims 1 to 9.

Images