A method and system for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons in humans.

Abstract

This invention discloses a method and system for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons (PAHs) in the human body. This invention transforms a physical model into a mathematical model, solving it by directly drawing symbols in software and inputting parameters, unlike traditional code-based solutions. By inputting parameter values ​​and external exposure levels, it predicts the temporal changes in PAH concentrations in human tissues in children and adults. Compared to the traditional code-based solution process, this invention simplifies the solution method, allowing prediction of the temporal trend of PAH concentration changes and total accumulation in human tissues after entry into the human body simply by inputting parameters and residual amounts into the module.

Description

Technical Field

[0001] This invention relates to the fields of computational toxicology and toxicological data processing, specifically to a method and system for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons in humans. Background Technology

[0002] Polycyclic aromatic hydrocarbons (PAHs) are a class of organic compounds with unique structures and properties. They play important roles in numerous fields and have complex environmental impacts. Their chemical structure consists of two or more benzene rings linked in a fused ring configuration, without other functional groups. PAHs exhibit varying degrees of toxicity, with some possessing strong carcinogenicity, mutagenicity, and reproductive toxicity. They are persistent pollutants of key concern to the United Nations Environment Programme (UNEP), primarily entering the human body through respiratory deposition, skin adsorption, and dietary exposure.

[0003] Toxicokinetics is the science that studies the changes in the absorption, distribution, metabolism, and excretion (ADME) of exogenous chemicals (such as poisons and drugs) in the body over time. Toxicokinetic models use mathematical methods to describe the quantitative relationships of these processes. Currently, physiological toxicokinetic models (PBTK) are widely used to simulate the ADME processes of organic chemicals in fish, rodents, and humans. PBTK models simulate the concentration distribution of target compounds in each compartment by dividing the body into several physiologically significant compartments (such as the liver, lungs, arteries, and veins), thereby constructing a series of differential equations with mass balance.

[0004] However, existing studies have focused on respiratory / skin contact exposure, with insufficient research on the kinetic characteristics of dietary exposure (such as single-peak concentration changes), and have not optimized parameters for dietary exposure scenarios, resulting in insufficient model prediction accuracy. Furthermore, the metabolic differences among different populations (such as clearance rate and tissue mass-volume ratio) have not been systematically analyzed, leading to unclear mechanisms in susceptible populations during risk assessment. Summary of the Invention

[0005] To address these issues, this invention provides a method and system for constructing a human polycyclic aromatic hydrocarbon (PAH) physiological toxicokinetic kinetic model, thereby resolving the technical problems of existing technologies such as insufficient model prediction accuracy due to the lack of parameter optimization for dietary exposure scenarios, lack of systematic analysis of metabolic differences among different populations, and unclear mechanisms in susceptible populations.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] According to a first aspect of the present invention, a method for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons in humans is provided, the method being applied to a symbolic modeling unit, comprising:

[0008] Obtain user requirements and select at least one polycyclic aromatic hydrocarbon and parameters based on those requirements, and limit the range accordingly;

[0009] A toxicokinetic kinetic model is constructed using a symbolic modeling unit, and the selected parameters and limits are input to generate kinetic process equations. The input, compartmentalization, and metabolic pathway of polycyclic aromatic hydrocarbons are defined through a graphical interface.

[0010] The compartments are connected and metabolic pathways are integrated using inhibition coefficients, Michaelis equations, and co-metabolic parameters. Then, machine learning algorithms are used to optimize the metabolic rate constant, generating a trained toxicokinetic model.

[0011] The selected polycyclic aromatic hydrocarbons and their corresponding parameters were input into the toxicokinetics model to obtain the time-varying trend of polycyclic aromatic hydrocarbon concentrations in various tissues and the total enrichment data after the polycyclic aromatic hydrocarbons entered the human body.

[0012] The selected parameters include physicochemical properties, metabolic parameters, and physiological parameters.

[0013] Furthermore, the physicochemical properties include molecular weight, lipid-water partition coefficient, and dissociation constant;

[0014] The molecular weight is defined as 128–302 g / mol, the lipid-water partition coefficient is defined as logKow 4.5–6.5, and the dissociation constant is defined as pKa 0–14.

[0015] The metabolic parameters were as follows: stomach 7.5, 13.5; lungs 0.0216, 0.0226; heart 0.1056, 0.2722; liver 0.0352, 0.0287; kidneys 0.1709, 0.1029; fat 0.05, 0.05; muscle 0.1954, 0.0915; skin 0.0626, 0.02437.

[0016] The unit of the metabolic parameters is L / h;

[0017] The physiological parameter is a volume fraction, specifically including:

[0018] The mass-volume fractions were as follows: stomach 0.0832, 0.1014; lungs 0.0225, 0.0347; heart 0.0051, 0.0076; liver 0.0144, 0.0167; kidneys 0.0046, 0.0062; fat 0.184, 0.2148; muscle 0.4576, 0.3402; skin 0.0553, 0.0403.

[0019] The unit of the mass volume fraction is L / BW.

[0020] Furthermore, the equations for generating the kinetic process are as follows:

[0021] The distribution of polycyclic aromatic hydrocarbons in tissues such as blood, liver, and adipose tissue is described using mass balance equations.

[0022] The calculation formula for polycyclic aromatic hydrocarbons after entering each compartment is as follows:

[0023]

[0024] Wherein, Elimination is the clearance process of a substance in the central compartment, KLI is the renal and non-renal clearance rate, C is the central compartment concentration, Mc is the central compartment mass, and Vc represents the central compartment volume;

[0025] The calculation expression for polycyclic aromatic hydrocarbons during the distribution process is as follows:

[0026] Qi_Flow = Qi × (C - Ai)

[0027] Where Qi represents the clearance rate of each compartment, and Ai represents the concentration of each compartment;

[0028] The calculation expression for the kinetic process equation is as follows:

[0029]

[0030] Where +Dose represents the dietary intake protocol, Mi represents the concentration in each compartment, and Vi represents the volume in each compartment.

[0031] Furthermore, the metabolic pathway integration includes a competitive inhibition model and a mixed exposure simulation, wherein the competitive inhibition model includes inhibition coefficients determined based on in vitro experiments and enzymatic reactions described using the Michaelis-Menten equation;

[0032] The mixed exposure simulation integrates the competitive metabolism of polycyclic aromatic hydrocarbon mixtures by introducing synergistic metabolic parameters with an inhibition intensity coefficient of 0.5–2.0.

[0033] Furthermore, the machine learning algorithm is one of particle swarm optimization, grid search, random forest, and support vector machine;

[0034] The parameters to be optimized include the learning rate, the number of iterations, and the population size.

[0035] The learning rate ranges from 0.001 to 0.1, the number of iterations ranges from 100 to 1000, and the population size ranges from 20 to 100.

[0036] Furthermore, during parameter optimization, a probability distribution model of physiological parameters related to age, gender, and BMI is established, and the parameters of a specific population are optimized using the probability distribution model.

[0037] Furthermore, the method also includes:

[0038] The data on the temporal variation trend of polycyclic aromatic hydrocarbons (PAHs) concentration in various tissues after entering the human body and the total enrichment amount were compared with experimental data to verify whether the verification threshold was exceeded and to obtain data verification results.

[0039] Adjust the parameter range and continuously observe the changing trend of the data verification results, then optimize the parameter range based on the changing trend of the data verification results.

[0040] Furthermore, the preferred criteria for data validation are AUC prediction error ≤ 2 times the measured value and Cmax prediction error ≤ 3 times the measured value, and the parameter sensitivity and model stability are verified through Monte Carlo simulation.

[0041] According to a second aspect of the present invention, a system for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons in humans is provided, the system comprising:

[0042] A polycyclic aromatic hydrocarbon (PAH) selection module is used to obtain user requirements and select at least one PAH and parameters and limit the range according to the user requirements.

[0043] The modeling module is used to construct toxicokinetics models through symbolic modeling units and input the selected parameters and limits to generate kinetic equations. It defines the input, compartmentalization and metabolic pathways of polycyclic aromatic hydrocarbons through a graphical interface.

[0044] The optimization module connects the compartments and integrates metabolic pathways using inhibition coefficients, Michaelis equations, and co-metabolic parameters. Then, it uses machine learning algorithms to optimize the metabolic rate constant and generate a trained toxicokinetic model.

[0045] The model application module is used to input the selected polycyclic aromatic hydrocarbons and corresponding parameters into the toxicokinetics model to obtain the time-varying trend of polycyclic aromatic hydrocarbon concentrations in various tissues and the total enrichment data after polycyclic aromatic hydrocarbons enter the human body.

[0046] The selected parameters include physicochemical properties, metabolic parameters, and physiological parameters.

[0047] The embodiments of the present invention have the following advantages:

[0048] This invention transforms a physical model into a mathematical model, solving the problem by directly drawing symbols in the software and inputting parameters, unlike traditional code-based solutions. It predicts the temporal changes in polycyclic aromatic hydrocarbon (PAH) tissue concentrations in children and adults by inputting parameter values ​​and external exposure levels. Compared to the traditional code-based solution process, this invention simplifies the solution method, allowing prediction of the temporal trend of PAH tissue concentration changes and total enrichment after PAHs enter the human body simply by inputting parameters and residual amounts into the module. Attached Figure Description

[0049] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0050] The structures, proportions, sizes, etc. illustrated in this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed herein, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0051] Figure 1 A schematic diagram of the logical structure of a system for constructing a physiological toxicokinetics model of polycyclic aromatic hydrocarbons in the human body, provided in an embodiment of the present invention;

[0052] Figure 2 A flowchart illustrating a method for constructing a physiological toxicokinetics model of polycyclic aromatic hydrocarbons in the human body, provided in an embodiment of the present invention;

[0053] Figure 3 A schematic diagram of the physical model for constructing a physiological toxicokinetics model of polycyclic aromatic hydrocarbons in the human body, as provided in an embodiment of the present invention;

[0054] Figure 4 This is a schematic diagram illustrating the dynamic changes in concentrations of polycyclic aromatic hydrocarbons (PAHs) in various tissues of children, as provided in an embodiment of the present invention for constructing a physiological toxicokinetic kinetic model of human polycyclic aromatic hydrocarbons.

[0055] Figure 5 This is a schematic diagram of animal experiment verification of the prediction results in the method for constructing a human polycyclic aromatic hydrocarbon physiological toxicokinetics model provided in an embodiment of the present invention.

[0056] Figure 6 is a schematic diagram illustrating the prediction of time-varying concentrations of polycyclic aromatic hydrocarbons (PAHs) in human tissues in a method for constructing a physiological toxicokinetics model of PAHs provided in an embodiment of the present invention. Detailed Implementation

[0057] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. 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.

[0058] Polycyclic aromatic hydrocarbons (PAHs) are persistent pollutants of key concern to the United Nations Environment Programme (UNEP). They enter the human body through respiratory deposition, skin adsorption, and dietary exposure. With the rapid development of my country's population and industrialization and urbanization, the amount of PAHs emitted by human activities has increased, becoming a significant factor contributing to environmental pollution and posing certain health risks to Chinese residents. Currently, research on the toxicokinetics of PAHs exposure in humans through diet and the methods for assessing the risks associated with these mechanisms is still in the exploratory and developmental stage in China. This invention develops a PAH toxicokinetics model to provide technical support and scientific basis for improving the effectiveness of PAH dietary risk control.

[0059] Physiological toxicokinetic (PBTK) models are now widely used to simulate the absorption, distribution, metabolism, and excretion (ADME) of organic chemicals in fish, rodents, and humans. PBTK models simulate the concentration distribution of target compounds in each compartment by dividing the body into physiologically significant compartments (such as the liver, lungs, arteries, and veins), thus constructing a series of differential equations with mass balance. Many researchers use PBTK models to simulate internal exposure to various environmental pollutants, including volatile organic compounds, industrial chemicals, and endocrine disruptors. The development of toxicokinetic models for PAHs is still in the exploratory stage.

[0060] To address the aforementioned technical issues, such as insufficient model prediction accuracy due to lack of parameter optimization for dietary exposure scenarios, lack of systematic analysis of metabolic differences between children and adults, and unclear mechanisms in susceptible populations.

[0061] refer to Figure 1 This invention discloses a system for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons (PAHs) in the human body. The system includes: a PAH selection module 1; a modeling module 2; an optimization module 3; and a model application module 4.

[0062] Corresponding to the aforementioned system for constructing a human polycyclic aromatic hydrocarbon (PAH) physiological toxicokinetic model, this invention also discloses a method for constructing such a model. The following details the method for constructing a human PAH physiological toxicokinetic model disclosed in this invention, in conjunction with the aforementioned system for constructing such a model.

[0063] refer to Figure 2 This invention discloses a method for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons in the human body. The method is applied to a symbolic modeling unit and includes:

[0064] Obtain user requirements and select at least one polycyclic aromatic hydrocarbon and parameters based on those requirements, and limit the range accordingly;

[0065] A toxicokinetic kinetic model is constructed using a symbolic modeling unit, and the selected parameters and limits are input to generate kinetic process equations. The input, compartmentalization, and metabolic pathway of polycyclic aromatic hydrocarbons are defined through a graphical interface.

[0066] The compartments are connected and metabolic pathways are integrated using inhibition coefficients, Michaelis equations, and co-metabolic parameters. Then, machine learning algorithms are used to optimize the metabolic rate constant, generating a trained toxicokinetic model.

[0067] The selected polycyclic aromatic hydrocarbons and their corresponding parameters were input into the toxicokinetics model to obtain the time-varying trend of polycyclic aromatic hydrocarbon concentrations in various tissues and the total enrichment data after the polycyclic aromatic hydrocarbons entered the human body.

[0068] The selected parameters include physicochemical properties, metabolic parameters, and physiological parameters.

[0069] Further, the physicochemical properties include molecular weight, lipid-water partition coefficient, and dissociation constant; wherein the molecular weight is limited to a range of 128–302 g / mol, the lipid-water partition coefficient is limited to a range of logKow 4.5–6.5, and the dissociation constant is limited to a range of pKa 0–14; the metabolic parameters are: stomach 7.5, 13.5, lung 0.0216, 0.0226, heart 0.1056, 0.2722, liver 0.0352, 0.0287, kidney 0.1709, 0.1029, fat 0.05, 0.05, muscle 0.1954, 0.0915, and skin 0.0626, 0.02437.

[0070] The unit of the metabolic parameters is L / h.

[0071] The physiological parameters are volume fractions, specifically including the following volume fractions: stomach 0.0832, 0.1014; lungs 0.0225, 0.0347; heart 0.0051, 0.0076; liver 0.0144, 0.0167; kidneys 0.0046, 0.0062; fat 0.184, 0.2148; muscle 0.4576, 0.3402; skin 0.0553, 0.0403; wherein the unit of the volume fraction is L / BW.

[0072] Parameter selection and range limitation:

[0073] (1) Types of PAHs: Select at least one of 18 common PAHs, including naphthalene, phenanthrene, pyrene, benzo(a)pyrene (BaP), and dibenzo(a,h)anthracene (DBA), preferably a combination including BaP, DBA, and fluoranthracene (Fla);

[0074] (2) Physicochemical properties: including molecular weight (128~302g / mol), lipid-water partition coefficient (logKow3.0~7.5), dissociation constant (pKa0~14), preferably logKow4.5~6.5;

[0075] (3) Metabolic parameters (L / h): Stomach 7.5, 13.5; Lung 0.0216, 0.0226; Heart 0.1056, 0.2722; Liver 0.0352, 0.0287; Kidney 0.1709, 0.1029; Fat 0.05, 0.05; Muscle 0.1954, 0.0915; Skin 0.0626, 0.02437.

[0076] (4) Physiological parameters: mass-volume fraction (L / BW) Stomach 0.0832, 0.1014; Lung 0.0225, 0.0347; Heart 0.0051, 0.0076; Liver 0.0144, 0.0167; Kidney 0.0046, 0.0062; Fat 0.184, 0.2148; Muscle 0.4576, 0.3402; Skin 0.0553, 0.0403.

[0077] Furthermore, kinetic equations are generated, including: describing the distribution of polycyclic aromatic hydrocarbons in tissues such as blood, liver, and fat through mass balance equations;

[0078] The calculation formula for polycyclic aromatic hydrocarbons after entering each compartment is as follows:

[0079]

[0080] Wherein, Elimination is the clearance process of a substance in the central compartment, KLI is the renal and non-renal clearance rate, C is the central compartment concentration, Mc is the central compartment mass, and Vc represents the central compartment volume;

[0081] The distribution process represents the kinetics of a substance flowing from the central compartment to other compartments, and is proportional to the concentration difference. The calculation expression for polycyclic aromatic hydrocarbons (PAHs) during the distribution process is as follows:

[0082] Qi_Flow = Qi × (C - Ai)

[0083] Where Qi represents the clearance rate of each compartment, and Ai represents the concentration of each compartment;

[0084] The calculation expression for the kinetic process equation is as follows:

[0085]

[0086] Where +Dose represents the dietary intake protocol, Mi represents the concentration in each compartment, and Vi represents the volume in each compartment.

[0087] In this process, refer to Figure 3 , Figure 3 This diagram illustrates the seven compartments of the human body exposed to PAHs, simulating the process by which PAHs enter the gastrointestinal tract via food intake, undergo initial metabolism in the stomach, and are excreted in feces. The remainder is distributed throughout the body's organs via the bloodstream. The model includes the following tissues and organs: stomach (central compartment), lungs, heart, kidneys (excreting PAHs and their metabolites), liver (the only organ responsible for metabolic functions), muscle, skin, and adipose tissue. Each compartment is interconnected via arterial and venous blood circulation pathways, reflecting the complex network of blood flow within the body.

[0088] Furthermore, the metabolic pathway integration includes a competitive inhibition model and a mixed exposure simulation, wherein the competitive inhibition model includes an inhibition coefficient determined based on in vitro experiments and an enzymatic reaction described using the Michaelis-Menten equation; the mixed exposure simulation integrates the competitive metabolism of a mixture of polycyclic aromatic hydrocarbons by introducing synergistic metabolic parameters, with an inhibition strength coefficient of 0.5 to 2.0.

[0089] Metabolic pathway integration includes competitive inhibition models and mixed exposure simulations:

[0090] 1) Competitive inhibition model: based on the inhibition coefficient determined in vitro (e.g., the K-value of BaP on DBC metabolism). i The enzymatic reaction was described using the Michaelis-Menten equation (at a concentration of 0.061 μM).

[0091] 2) Mixed exposure simulation: For PAH mixtures (such as Supermix-10), synergistic metabolic parameters (inhibition intensity coefficient 0.5 to 2.0) are introduced.

[0092] Furthermore, the machine learning algorithm is one of particle swarm optimization, grid search, random forest, and support vector machine; the parameter optimization range includes learning rate, number of iterations, and population size; wherein the learning rate ranges from 0.001 to 0.1, the number of iterations ranges from 100 to 1000, and the population size ranges from 20 to 100.

[0093] Machine learning algorithms: Particle swarm optimization (PSO) or grid search (GridSearchCV) are used to optimize the metabolic rate constant (e.g., the Vmax of CYP1A1 is optimized from 0.1 to 100 pmol / min / mg protein).

[0094] The machine learning algorithm can be replaced by random forest or support vector machine, and the parameter optimization range includes:

[0095] Learning rate (0.001–0.1), number of iterations (100–1000), population size (20–100)

[0096] Furthermore, during parameter optimization, a probability distribution model of physiological parameters related to age, gender, and BMI is established, and the parameters of a specific population are optimized using the probability distribution model.

[0097] Population variability: Considering age (18–80 years), sex (male / female), and BMI (18.5–30 kg / m²). 2 To investigate the impact of physiological parameters, a probability distribution model was established.

[0098] Furthermore, the method also includes: comparing the time-varying trend of polycyclic aromatic hydrocarbon concentrations in various tissues and the total enrichment data with experimental data after polycyclic aromatic hydrocarbons enter the human body to verify whether the verification threshold is exceeded, and obtaining data verification results; adjusting the parameter range and continuously observing the changing trend of the data verification results, and optimizing the parameter range based on the changing trend of the data verification results.

[0099] Furthermore, the preferred criteria for data validation are AUC prediction error ≤ 2 times the measured value and Cmax prediction error ≤ 3 times the measured value, and the parameter sensitivity and model stability are verified through Monte Carlo simulation.

[0100] The validation process also requires animal experiments, specifically including validation based on in vivo toxicokinetics data of rats 48 hours after oral administration of PAHs.

[0101] The PBTK model constructed using the embodiments of the present invention simulated the dynamic changes in the concentration of PAHs in various tissues in adults and children after dietary exposure to PAHs. Figure 4 To simulate the time-varying trend of PAH concentrations in various tissues of children. Figure 5 A graph showing the time-varying trend of PAH concentrations in some tissues of rats over 48 hours after oral administration.

[0102] like Figure 4 and Figure 5 As shown, taking children as an example, the concentration of PAHs in various tissues exhibits a typical "single-peak" characteristic over time, meaning that the concentration rises rapidly to a peak after exposure and then gradually decreases. This pattern is highly consistent with other toxicokinetic studies in rats. Specifically, after absorption through the digestive tract, PAHs reach maximum concentrations in most tissues within 0-10 hours, mainly due to the rapid absorption of PAHs in the gastrointestinal tract and their distribution in the bloodstream.

[0103] The application scenarios of this invention are exposure assessment and health risk assessment:

[0104] Exposure assessment: Predicted concentrations in the human body after dietary intake (9.75, 6.31, 6.04 μg / kgbw);

[0105] Health risk assessment: Combine toxicity endpoints (such as the carcinogenic risk factor of BaP, 7.75 mg / kg / day) to output safety thresholds.

[0106] It should be noted that the PAHs can be replaced with other polycyclic aromatic hydrocarbons (such as benzo(a)pyrene, indo(1,2,3-cd)pyrene), and the numerical parameter range can be adjusted as follows:

[0107] Concentration range: Environmental exposure concentration (0.01~10μg / m³) 3 High-dose exposure (≥100mg);

[0108] Temperature range: in vitro metabolic experiment temperature (37±0.5℃), in vivo physiological temperature (36.1~37℃).

[0109] Furthermore, the method proposed in this embodiment of the invention can be integrated into a software platform (such as GastroPlus, Simcyp) and supports user-defined parameter ranges (such as metabolic enzyme activity ±30%).

[0110] Berkeley Madonna is a fast, general-purpose differential equation solver that can be used to build complex mathematical models. It allows for direct plotting of results after symbolic modeling, and the software enables direct modification of parameter values ​​via a parameter window while automatically generating equations. It can also solve ordinary differential equations, difference equations, and roots of multidimensional transcendental algebraic equations by writing code.

[0111] Commonly used functions include Euler (first order), Runge-Kutta (second and fourth order), adaptive step size (fourth order Runge-Kutta), the StiffODE solver (Rosenbrock), and custom DT (decode your own equations with adjustable step sizes). It allows for stochastic modeling using methods such as the Gillespie algorithm. It is a software that integrates stochastic modeling and code scripting.

[0112] Modeling parameters are usually derived from a database. The basic structure of the specific model during the modeling process includes: well-perfused tissues, including organs with high blood flow perfusion rates such as the brain, liver, and kidneys; and poorly perfused tissues, including tissues with low blood flow perfusion rates such as muscles and skin.

[0113] This invention, for the first time, incorporates the competitive metabolism of PAH mixtures (e.g., BaP's inhibition of DBC) into the PBTK model, with inhibition coefficients ranging from 0.01 to 100 μM. Considering population variability, age, sex, and BMI-related physiological parameter distribution models are established (e.g., liver volume variation coefficient of 10-20%). Optimization is achieved through machine learning: metabolic parameters are optimized using the PSO algorithm (e.g., Vmax optimization accuracy of CYP1A1 is ±5%). Validation thresholds of ≤2-fold IVIVC error and ≤3-fold error in animal experiments are proposed for standard model validation.

[0114] Figure 6 shows a comparison of PAH concentrations in simulated tissues from different populations. Figure 6a For adult males Figure 6b For adult women, Figure 6c For children.

[0115] Please refer to Figure 6. Using the PBTK model proposed in this embodiment of the invention, the changes in tissue concentration over time in children and adults are predicted by inputting parameter values ​​and external exposure amounts. It can be seen that urban residents rapidly reach the highest tissue concentration levels at the initial stage of exposure. The maximum concentrations in various tissues in children are as follows: liver (2.06 ng / L) > muscle (1.80 ng / L) > fat (1.46 ng / L) > heart (0.68 ng / L) > kidney (0.40 ng / L) > skin (0.37 ng / L) > lung (0.34 ng / L); the tissue concentrations in adult men are... The concentrations of nitric oxide in the body were as follows: muscle (1.57 ng / L) > liver (1.02 ng / L) > fat (0.53 ng / L) > skin (0.35 ng / L) > kidney (0.23 ng / L) > heart (0.20 ng / L) > lung (0.12 ng / L); the concentrations in adult female tissues were as follows: muscle (1.52 ng / L) > liver (0.97 ng / L) > fat (0.51 ng / L) > skin (0.34 ng / L) > kidney (0.22 ng / L) > heart (0.19 ng / L) > lung (0.12 ng / L).

[0116] Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A method for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons in humans, characterized in that, The method is applied to a symbolic modeling unit, which includes: Obtain user requirements and select at least one polycyclic aromatic hydrocarbon and parameters based on those requirements, and limit the range accordingly; A toxicokinetic kinetic model is constructed using a symbolic modeling unit, and the selected parameters and limits are input to generate kinetic process equations. The input, compartmentalization, and metabolic pathway of polycyclic aromatic hydrocarbons are defined through a graphical interface. The compartments are connected and metabolic pathways are integrated using inhibition coefficients, Michaelis equations, and co-metabolic parameters. Then, machine learning algorithms are used to optimize the metabolic rate constant, generating a trained toxicokinetic model. The selected polycyclic aromatic hydrocarbons and their corresponding parameters were input into the toxicokinetics model to obtain the time-varying trend of polycyclic aromatic hydrocarbon concentrations in various tissues and the total enrichment data after the polycyclic aromatic hydrocarbons entered the human body. The selected parameters include physicochemical properties, metabolic parameters, and physiological parameters.

2. The method for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons in humans as described in claim 1, characterized in that, The physicochemical properties include molecular weight, lipid-water partition coefficient, and dissociation constant; The molecular weight is defined as 128–302 g / mol, the lipid-water partition coefficient is defined as logKow 4.5–6.5, and the dissociation constant is defined as pKa 0–14. The metabolic parameters were as follows: stomach 7.5, 13.5; lungs 0.0216, 0.0226; heart 0.1056, 0.2722; liver 0.0352, 0.0287; kidneys 0.1709, 0.1029; fat 0.05, 0.05; muscle 0.1954, 0.0915; skin 0.0626, 0.02437. The units of the metabolic parameters are L / h; The physiological parameter is a volume fraction, specifically including: The mass-volume fractions were as follows: stomach 0.0832, 0.1014; lungs 0.0225, 0.0347; heart 0.0051, 0.0076; liver 0.0144, 0.0167; kidneys 0.0046, 0.0062; fat 0.184, 0.2148; muscle 0.4576, 0.3402; skin 0.0553, 0.0403. The unit of the mass volume fraction is L / BW.

3. The method for constructing a physiological toxicokinetic kinetic model of human polycyclic aromatic hydrocarbons as described in claim 2, characterized in that, Equations for generating kinetic processes include: The distribution of polycyclic aromatic hydrocarbons in blood, liver, and adipose tissue is described using mass balance equations; The calculation formula for polycyclic aromatic hydrocarbons after entering each compartment is as follows: Wherein, Elimination is the clearance process of a substance in the central compartment, KLI is the renal and non-renal clearance rate, C is the central compartment concentration, Mc is the central compartment mass, and Vc represents the central compartment volume; The calculation expression for polycyclic aromatic hydrocarbons during the distribution process is as follows: Qi_Flow = Qi × (C - Ai) Where Qi represents the clearance rate of each compartment, and Ai represents the concentration of each compartment; The calculation expression for the kinetic process equation is as follows: Where +Dose represents the dietary intake protocol, Mi represents the concentration in each compartment, and Vi represents the volume in each compartment.

4. The method for constructing a physiological toxicokinetic kinetic model of human polycyclic aromatic hydrocarbons as described in claim 3, characterized in that, The metabolic pathway integration includes competitive inhibition models and mixed exposure simulations, wherein the competitive inhibition models include inhibition coefficients based on in vitro experimental determinations and enzymatic reactions described using the Michaelis-Menten equation; The mixed exposure simulation integrates the competitive metabolism of polycyclic aromatic hydrocarbon mixtures by introducing synergistic metabolic parameters with an inhibition intensity coefficient of 0.5–2.

0.

5. The method for constructing a physiological toxicokinetic kinetic model of human polycyclic aromatic hydrocarbons as described in claim 4, characterized in that, The machine learning algorithm is one of particle swarm optimization, grid search, random forest, and support vector machine; The parameters to be optimized include the learning rate, the number of iterations, and the population size. The learning rate ranges from 0.001 to 0.1, the number of iterations ranges from 100 to 1000, and the population size ranges from 20 to 100.

6. The method for constructing a physiological toxicokinetic kinetic model of human polycyclic aromatic hydrocarbons as described in claim 5, characterized in that, During parameter optimization, a probability distribution model of physiological parameters related to age, gender, and BMI is established, and the parameters of a specific population are optimized using the probability distribution model.

7. The method for constructing a physiological toxicokinetic kinetic model of human polycyclic aromatic hydrocarbons as described in claim 6, characterized in that, The method further includes: The data on the temporal variation trend of polycyclic aromatic hydrocarbons (PAHs) concentration in various tissues after entering the human body and the total enrichment amount were compared with experimental data to verify whether the verification threshold was exceeded and to obtain data verification results. Adjust the parameter range and continuously observe the changing trend of the data verification results, then optimize the parameter range based on the changing trend of the data verification results.

8. The method for constructing a physiological toxicokinetic kinetic model of human polycyclic aromatic hydrocarbons as described in claim 7, characterized in that, The preferred criteria for data validation are AUC prediction error ≤ 2 times the measured value and Cmax prediction error ≤ 3 times the measured value. The parameter sensitivity and model stability are verified by Monte Carlo simulation.

9. A system for constructing a physiological toxicokinetic kinetic model of polycyclic aromatic hydrocarbons in humans, characterized in that, The system includes: A polycyclic aromatic hydrocarbon (PAH) selection module is used to obtain user requirements and select at least one PAH and parameters and limit the range according to the user requirements. The modeling module is used to construct toxicokinetics models through symbolic modeling units and input the selected parameters and limits to generate kinetic equations. It defines the input, compartmentalization and metabolic pathways of polycyclic aromatic hydrocarbons through a graphical interface. The optimization module connects the compartments and integrates metabolic pathways using inhibition coefficients, Michaelis equations, and co-metabolic parameters. Then, it uses machine learning algorithms to optimize the metabolic rate constant and generate a trained toxicokinetic model. The model application module is used to input the selected polycyclic aromatic hydrocarbons and corresponding parameters into the toxicokinetics model to obtain the time-varying trend of polycyclic aromatic hydrocarbon concentrations in various tissues and the total enrichment data after polycyclic aromatic hydrocarbons enter the human body. The selected parameters include physicochemical properties, metabolic parameters, and physiological parameters.

Images