A method and system for constructing a digital twin model of a zebrafish toxicity effect

Abstract

The application discloses a digital twin model construction method and system for zebrafish toxicity effect, and relates to the technical field of environmental toxicology simulation. Zebrafish physiological parameters, behavior characteristics, environmental exposure parameters and toxicity endpoint data are acquired to construct a modeling database; the data is preprocessed; an energy metabolism and behavior coupling model is constructed, the behavior of zebrafish is divided into multiple states and corresponding metabolic rates are set, a circadian rhythm mechanism is introduced to perform energy balance calculation; a toxicokinetic-toxicodynamic model is constructed, a one-compartment model is used to describe the absorption and discharge process of pollutants in the zebrafish body, and a dynamic correction mechanism of energy state on absorption rate constant and discharge rate constant is established; when the concentration in the body reaches a toxicity response threshold, a behavior response is triggered, and energy consumption is recalculated; and a zebrafish digital twin simulation model and a visual interactive interface are constructed based on the above model. The application realizes dynamic simulation of the pollutant exposure process and three-dimensional visual expression of the behavior response.

Description

Technical Field

[0001] This invention relates to the field of environmental toxicology simulation technology, and in particular to a method and system for constructing a digital twin model of zebrafish toxicity effects. Background Technology

[0002] With the continuous advancement of industrialization and urbanization, large amounts of pollutants, such as heavy metals, are continuously released into the aquatic environment, where they migrate, transform, and accumulate, posing a long-term threat to aquatic ecosystems and biosecurity. Among numerous model organisms, zebrafish (Danio rerio) are widely used in environmental toxicology and pharmacology research due to their small size, short reproductive cycle, transparent embryos, ease of observing developmental processes, and high homology with humans in genetic and metabolic pathways. Existing studies typically assess the toxic effects of heavy metals by controlling heavy metal concentrations and exposure time, observing the survival rate, behavioral changes, and physiological responses of zebrafish.

[0003] However, traditional toxicity testing methods have significant limitations. On the one hand, these experiments rely on large numbers of biological samples, have long experimental cycles, are costly, and the experimental conditions are difficult to standardize completely, resulting in limited reproducibility of results. On the other hand, the experimental data are mostly observations at discrete time points, making it difficult to continuously characterize the dynamic changes of heavy metals in organisms, especially in the kinetic processes such as absorption, accumulation, and excretion. The lack of systematic and visual representation means that the understanding of the toxic mechanism remains at the characterization level, lacking in-depth characterization of the process and the system.

[0004] To address these shortcomings, some studies have attempted to introduce toxikinetic-toxicodynamic (TK-TD) models, which use mathematical equations to describe the migration, transformation, and toxic effects of heavy metals within organisms. These methods have improved the analytical capabilities regarding toxicity mechanisms to some extent. However, existing models often focus on parameter fitting and numerical calculations, typically separating them from the behavioral characteristics and energy metabolism processes of individual organisms. They lack a comprehensive characterization of the overall state of the organism, making it difficult to achieve intuitive visualization and interactive analysis.

[0005] In recent years, digital twin technology, as a simulation method that integrates physical entities and virtual models, has been able to achieve dynamic mapping and process reproduction of real systems, providing a new technical path for modeling complex biological processes. However, in the field of aquatic toxicology, existing research has mostly focused on single mathematical models or two-dimensional data analysis, and there is a lack of digital twin methods that combine zebrafish physiological processes, energy metabolism mechanisms, toxicokinetics-toxicodynamics models with three-dimensional behavioral simulation.

[0006] To address the aforementioned issues, how to propose a digital twin method based on toxicokinetics to simulate the toxic effects of zebrafish, enabling dynamic simulation of heavy metal exposure processes and visualization of behavioral responses, thereby providing a more intuitive and efficient technical means for toxicological research, has become a pressing technical problem for those skilled in the art. Summary of the Invention

[0007] In view of this, the present invention provides a method and system for constructing a digital twin model of zebrafish toxicity effects. It addresses the shortcomings of existing in vivo toxicity experiments, traditional TK-TD numerical models, and aquatic toxicology simulation techniques, and realizes closed-loop linkage simulation of energy metabolism, pollutant in vivo dynamics, and poisoning behavior in zebrafish under heavy metal exposure. The entire process of toxicity evolution is intuitively displayed through three-dimensional digital twins, reducing reliance on in vivo experiments and improving the ability to dynamically analyze toxicity mechanisms.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: A method for constructing a digital twin model of zebrafish toxicity effects includes the following steps: S1: Multi-source data acquisition and modeling database construction: Collect multi-dimensional toxicological data of zebrafish through public toxicity databases and academic literature, and construct a dedicated basic database for digital twins; S2: Data standardization preprocessing: Perform preprocessing on the raw data in the database to output a standard dataset suitable for modeling operations; S3: Construction of a coupling model of energy metabolism and behavior: Based on the physiological and behavioral characteristics of zebrafish, an energy balance equation is introduced to establish an energy balance, which is used to characterize the dynamic energy balance of zebrafish under different behavioral states; S4: Toxicity-toxicity kinetic model construction to characterize the dynamic changes of pollutants in zebrafish and establish a feedback mechanism between toxic effects and behavioral responses. S5: Digital Twin Simulation Model Development and Visual Interaction Setup: Construct a three-dimensional virtual entity model of a zebrafish, establish a mapping relationship between numerical calculation parameters and three-dimensional animation driving parameters, and develop a visual interactive interface that allows for customizable input conditions and real-time output of simulation results.

[0009] Optionally, the data collected by S1 in the above method includes: zebrafish physiological parameter data, including body weight W, body length L, standard metabolic rate SMR, maximum metabolic rate MMR, oxygen consumption rate and energy conversion efficiency. Behavioral characteristics data include: swimming speed v, behavioral time allocation ratio, and diurnal rhythm parameters; Environmental exposure parameters include: heavy metal concentration Cw, exposure time t, and water temperature T; Toxicity endpoint data include: median lethal concentration (LC50), survival rate, and behavioral abnormality indicators; The publicly available toxicity database is the ECOTOX database, and the academic literature data source is the Web of Science.

[0010] The above method, optionally, includes the following preprocessing techniques for S2: missing value cleaning, numerical imputation, category coding, and unit unification and standardization; the specific steps are as follows: S21: Handling missing values: Remove data records with a missing ratio exceeding a preset threshold; S22: Numerical data is processed uniformly; for interval data, the mean is used as a substitute, and missing values ​​are filled with the average value of similar features; for categorical data, a coding method is used for digital processing. S23: Standardize environmental factors and physiological parameters to avoid the influence of different dimensions on the model. The standardization formula is as follows:

[0011] in, It represents the data sequence number. i It is a positive integer. It is the first i 1 eigenvalue, It is the first i The average of the features, It is the first i The standard deviation (SD) of each feature Then it is the first i The eigenvalues ​​after normalization.

[0012] Optionally, in S3, the construction of the energy metabolism and behavior coupling model includes the following: Zebrafish behavior is divided into at least two behavioral states, and a corresponding metabolic rate is set for each behavioral state. Set energy intake parameters and calculate energy input. Calculate energy consumption based on the time allocation and corresponding metabolic rate of each behavioral state; The real-time energy state of zebrafish is determined based on the difference between energy input and energy consumption.

[0013] The above methods may optionally include the following behavioral states: cruising state, sprinting state, and stationary state. The metabolic rate corresponding to the cruising state is 2-3 times the standard metabolic rate (SMR), the metabolic rate corresponding to the sprint state is the maximum metabolic rate (MMR), and the metabolic rate corresponding to the resting state is the standard metabolic rate (SMR). By introducing a circadian rhythm mechanism, the 24-hour period is divided into a light period and a dark period, and the metabolic level is reduced during the dark period.

[0014] Optionally, the S4 toxicokinetics-toxicodynamics model uses a one-compartment model to describe the dynamic changes of pollutants in zebrafish.

[0015] in, This refers to the concentration in the body. For environmental concentration, The absorption rate constant is The excretion rate constant is given; the steady-state enrichment concentration in the body is:

[0016] Through parameters and Calculate bioaccumulation factors BCF : This refers to the concentration in the body. For environmental concentration.

[0017] Optionally, in the above method, the real-time energy state coefficient of the zebrafish can be calculated in S4:

[0018] in, Represents the energy state coefficient, real-time energy input. With real-time energy consumption when A value >0 indicates that the body is in a state of energy surplus. A value less than 0 indicates that the body is in a state of energy deficiency. Absorption rate constant and discharge rate constant Correct it as follows:

[0019] in, and These are the basic absorption rate constant and the basic excretion rate constant, respectively; α and β It is the energy regulation coefficient, used to characterize the degree of influence of energy state on the pollutant absorption and emission capacity.

[0020] Optionally, the feedback mechanism between the toxic effects and behavioral responses in S4 may include: When the concentration in the body When the preset toxicity response threshold is reached or exceeded, the zebrafish is determined to have entered a toxicity response state. In the state of toxic response, the time allocation ratio of each behavioral state is adjusted to reduce the proportion of high-energy-consuming behaviors and increase the proportion of low-energy-consuming behaviors. The energy consumption is recalculated based on the adjusted time allocation ratio, and the energy state parameters in the digital twin model are updated. The method for adjusting the time allocation ratio for activities is as follows:

[0021] in: The original parade time, The original sprint time, For the original rest time, The original parade time, The original sprint time, For the original rest time, T Total action time, λ and μ These represent the inhibition coefficients for cruising behavior and sprinting behavior, respectively. The energy consumption per unit time is recalculated based on the updated behavior time, and the expression is as follows:

[0022] in, SMR Indicates standard metabolic rate. MMR Indicates the maximum metabolic rate. This indicates the energy expenditure of specific dynamic processes after ingestion.

[0023] Optionally, in S5, the construction of the zebrafish digital twin simulation model includes the following: A three-dimensional virtual model of a zebrafish is constructed, and a mapping relationship is established between the model's calculation parameters and the virtual model's motion parameters, so that the virtual model's behavior response changes dynamically with the model's calculation results. Construct an interactive interface with adjustable parameters, where the adjustable parameters include at least one of pollutant type, exposure concentration, and exposure time; The behavioral responses of the virtual model include at least one of the following: changes in swimming speed, changes in tail wag frequency, and switching of behavioral states; The interactive interface is also used to output at least one of the following in real time: in vivo concentration change data, energy change data, and behavioral response results.

[0024] A digital twin model construction system for zebrafish toxicity effects, used to implement the method for constructing a digital twin model of zebrafish toxicity effects as described above, includes a database module, a data preprocessing module, an energy metabolism coupling calculation module, a toxicokinetics feedback calculation module, and a three-dimensional digital twin visualization simulation module connected in sequence. The database module collects multi-dimensional toxicological data of zebrafish from public toxicity databases and academic literature to build a dedicated basic database for digital twins. The data preprocessing module performs preprocessing on the raw data in the database and outputs a standard dataset adapted for modeling operations. The energy metabolism coupling calculation module, based on the physiological and behavioral characteristics of zebrafish, introduces an energy balance equation to establish energy balance, which is used to characterize the dynamic energy balance of zebrafish under different behavioral states. The toxicokinetics feedback calculation module is used to characterize the dynamic changes of pollutants in zebrafish and to establish a feedback mechanism between toxic effects and behavioral responses. 3D Digital Twin Visualization Simulation Module: Constructs a 3D virtual entity model of a zebrafish, establishes a mapping relationship between numerical calculation parameters and 3D animation driving parameters, and develops a visual interactive interface that allows for customizable input conditions and real-time output of simulation results.

[0025] As can be seen from the above technical solution, compared with the prior art, the present invention provides a method and system for constructing a digital twin model of zebrafish toxicity effects, which has the following beneficial effects: By coupling energy metabolism processes, behavioral characteristics, and toxicokinetics with a toxicokinetics-toxicodynamics model, dynamic simulation of heavy metal exposure in zebrafish was achieved, and the correlation between in vivo concentration changes and behavioral responses was established. Compared to traditional experimental methods, this invention can continuously reproduce the heavy metal exposure process in a computer environment, improving the dynamism and observability of toxicological studies. Furthermore, the results are visually presented through three-dimensional visualization, thereby enhancing the ability to express toxic processes while reducing experimental costs. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0027] Figure 1 This is a flowchart of a method for constructing a digital twin model of zebrafish toxicity effects disclosed in this invention; Figure 2This is a block diagram of the system structure for constructing a digital twin model of zebrafish toxicity effects disclosed in this invention. Detailed Implementation

[0028] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0029] In this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0030] See Figure 1 As shown, this invention discloses a method for constructing a digital twin model of zebrafish toxicity effects, comprising the following steps: S1: Multi-source data acquisition and modeling database construction: Collect multi-dimensional toxicological data of zebrafish through public toxicity databases and academic literature, and construct a dedicated basic database for digital twins; S2: Data standardization preprocessing: Perform preprocessing on the raw data in the database to output a standard dataset suitable for modeling operations; S3: Construction of a coupling model of energy metabolism and behavior: Based on the physiological and behavioral characteristics of zebrafish, an energy balance equation is introduced to establish an energy balance, which is used to characterize the dynamic energy balance of zebrafish under different behavioral states; S4: Toxicity-toxicity kinetic model construction to characterize the dynamic changes of pollutants in zebrafish and establish a feedback mechanism between toxic effects and behavioral responses. S5: Digital Twin Simulation Model Development and Visual Interaction Setup: Construct a three-dimensional virtual entity model of a zebrafish, establish a mapping relationship between numerical calculation parameters and three-dimensional animation driving parameters, and develop a visual interactive interface that allows for customizable input conditions and real-time output of simulation results.

[0031] Furthermore, the data collected by S1 includes: zebrafish physiological parameters, including body weight (W), body length (L), standard metabolic rate (SMR), maximum metabolic rate (MMR), oxygen consumption rate, and energy conversion efficiency. Behavioral data include: swimming speed v, time allocation of behavior (cruising proportion P1, sprinting proportion P2, stationary proportion P3), and circadian rhythm parameters (light duration Tlight, darkness duration Tdark). Environmental exposure parameters include: heavy metal concentration Cw, exposure time t, and water temperature T; Toxicity endpoint data include: median lethal concentration (LC50), survival rate, and behavioral abnormality indicators; The publicly available toxicity database is the ECOTOX database, and the academic literature data source is the Web of Science.

[0032] Specifically, the heavy metals include copper (Cu), cadmium (Cd), lead (Pb), and zinc (Zn); further, the preprocessing methods for S2 include: missing value cleaning, numerical imputation, category coding, and unit unification and standardization; the specific steps are as follows: S21: Handling missing values: Remove data records with a missing ratio exceeding a preset threshold to ensure data integrity and reliability; S22: Numerical data is processed uniformly. For interval data, the mean is used as a substitute, and missing values ​​are filled with the average value of similar features. For categorical data, a coding method is used for digital processing to meet the model input requirements. S23: Standardize environmental factors and physiological parameters to avoid the influence of different dimensions on the model. The standardization formula is as follows:

[0033] in, It represents the data sequence number. i It is a positive integer. It is the first i 1 eigenvalue, It is the first i The average of the features, It is the first i The standard deviation (SD) of each feature Then it is the first i The eigenvalues ​​after normalization.

[0034] After the above processing steps, the dataset meets the input standards for building a digital twin model.

[0035] Furthermore, in S3, the construction of the energy metabolism and behavior coupling model includes: Zebrafish behavior is divided into at least two behavioral states, and a corresponding metabolic rate is set for each behavioral state. Set energy intake parameters and calculate energy input. Calculate energy consumption based on the time allocation and corresponding metabolic rate of each behavioral state; The real-time energy state of zebrafish is determined based on the difference between energy input and energy consumption.

[0036] Furthermore, behavioral states include cruising, sprinting, and stationary states; The metabolic rate corresponding to the cruising state is 2-3 times the standard metabolic rate (SMR), the metabolic rate corresponding to the sprint state is the maximum metabolic rate (MMR), and the metabolic rate corresponding to the resting state is the standard metabolic rate (SMR). By introducing a circadian rhythm mechanism, the 24-hour period is divided into a light period and a dark period, during which the metabolic level is reduced; by integrating energy input and energy output, the dynamic energy change process of zebrafish is obtained.

[0037] Furthermore, S3 includes the following steps: S31, Feed Intake and Energy Input: Feed intake rate is set at 5% of body weight, feed energy value is 20 kJ / g, and absorption rate is 80%. Then:

[0038] S32 divides the behavior into: cruising (3-5 cm / s, accounting for about 95%), sprinting (about 20 cm / s, accounting for about 1%), and standing still (accounting for about 4%). S33, calculate metabolic rate: resting state is SMR, cruising state is 2-3 times SMR, and sprinting state is MMR. S34 is designed with a 14-hour light period and a 10-hour dark period, with the metabolic rate during the dark period being 5% to 20% of that during the light period. S35, calculates energy output, including basal metabolism, behavioral metabolism, and specific dynamics (SDA), i.e.:

[0039] Of these, specific dynamic effects (SDA) account for 15% of the energy intake.

[0040] Furthermore, the S4 toxicokinetics-toxicodynamics model uses a one-compartment model to describe the dynamic changes of pollutants in zebrafish:

[0041] in, This refers to the concentration in the body. For environmental concentration, The absorption rate constant is The excretion rate constant is given; the steady-state enrichment concentration in the body is:

[0042] Through parameters and Calculate bioaccumulation factors BCF : This refers to the concentration in the body. For environmental concentration.

[0043] Furthermore, the real-time energy state coefficient of the zebrafish is calculated in S4:

[0044] in, Represents the energy state coefficient, real-time energy input. With real-time energy consumption when A value >0 indicates that the body is in a state of energy surplus. A value less than 0 indicates that the body is in a state of energy deficiency. Absorption rate constant and discharge rate constant Correct it as follows:

[0045] in, and These are the basic absorption rate constant and the basic excretion rate constant, respectively; α and β It is the energy regulation coefficient, used to characterize the degree of influence of energy state on the pollutant absorption and emission capacity.

[0046] Furthermore, the feedback mechanism between the toxic effects of S4 and behavioral responses includes: When the concentration in the body When the preset toxicity response threshold is reached or exceeded, the zebrafish is determined to have entered a toxicity response state. In the state of toxic response, the time allocation ratio of each behavioral state is adjusted to reduce the proportion of high-energy-consuming behaviors and increase the proportion of low-energy-consuming behaviors. The energy consumption is recalculated based on the adjusted time allocation ratio, and the energy state parameters in the digital twin model are updated. The toxicity response threshold is expressed as the critical internal threshold (CIT), which is defined as the concentration of a substance that is in vivo when the following conditions are met:

[0047] At that time, the zebrafish were determined to have entered a state of toxic response, in which... This refers to the critical in vivo concentration determined based on experimental toxicity data of different pollutants.

[0048] The method for adjusting the time allocation ratio for activities is as follows:

[0049] in: The original parade time, The original sprint time, For the original rest time, To adjust the parade time, To adjust the sprint time, To adjust the rest time, T Total action time, λ and μ These represent the inhibition coefficients for cruising behavior and sprinting behavior, respectively. The energy consumption per unit time is recalculated based on the updated behavior time, and the expression is as follows:

[0050] in, SMR Indicates standard metabolic rate. MMR Indicates the maximum metabolic rate. This indicates the energy expenditure of specific dynamic processes after ingestion.

[0051] The recalculated energy consumption results are used to update the energy state parameters in the digital twin system and synchronously drive the swimming speed, tail wag frequency and behavioral state changes in the 3D animation, so as to achieve dynamic consistency between pollutant accumulation, physiological metabolic processes and behavioral responses.

[0052] Furthermore, in S5, the construction of the zebrafish digital twin simulation model includes: A three-dimensional virtual model of a zebrafish is constructed, and a mapping relationship is established between the model's calculation parameters and the virtual model's motion parameters, so that the virtual model's behavior response changes dynamically with the model's calculation results. Construct an interactive interface with adjustable parameters, where the adjustable parameters include at least one of pollutant type, exposure concentration, and exposure time; The behavioral responses of the virtual model include at least one of the following: changes in swimming speed, changes in tail wag frequency, and switching of behavioral states; The interactive interface is also used to output at least one of the following in real time: in vivo concentration change data, energy change data, and behavioral response results.

[0053] Specifically, in S5, the development of the digital twin model includes constructing a 3D simulation model of a zebrafish and performing motion binding, mapping the model calculation results with 3D animation parameters to achieve dynamic linkage between the virtual model and the computational model; at the same time, it constructs an interactive interface with adjustable parameters to set parameters such as heavy metal type, exposure concentration and exposure time, and outputs in-body concentration changes, energy changes and behavioral response results in real time.

[0054] A virtual model of zebrafish needs to be constructed using 3D modeling software (such as Blender) to realize processes such as behavioral animation and speed binding, circadian rhythm driving, and toxicity state driving abnormal behavior, and to establish the mapping relationship between model parameters and virtual behavior.

[0055] and Figure 1 Corresponding to the method shown, this invention also discloses a digital twin model construction system for zebrafish toxicity effects, used to achieve... Figure 1 The above describes a method for constructing a digital twin model of zebrafish toxicity effects. See the structural diagram below. Figure 2 As shown, it includes a database module, a data preprocessing module, an energy metabolism coupling calculation module, a toxicokinetics feedback calculation module, and a three-dimensional digital twin visualization simulation module, which are connected in sequence. The database module collects multi-dimensional toxicological data of zebrafish from public toxicity databases and academic literature to build a dedicated basic database for digital twins. The data preprocessing module performs preprocessing on the raw data in the database and outputs a standard dataset adapted for modeling operations. The energy metabolism coupling calculation module, based on the physiological and behavioral characteristics of zebrafish, introduces an energy balance equation to establish energy balance, which is used to characterize the dynamic energy balance of zebrafish under different behavioral states. The toxicokinetics feedback calculation module is used to characterize the dynamic changes of pollutants in zebrafish and to establish a feedback mechanism between toxic effects and behavioral responses. 3D Digital Twin Visualization Simulation Module: Constructs a 3D virtual entity model of a zebrafish, establishes a mapping relationship between numerical calculation parameters and 3D animation driving parameters, and develops a visual interactive interface that allows for customizable input conditions and real-time output of simulation results.

[0056] In one specific embodiment, the present invention constructs a virtual simulation model of zebrafish toxicity based on toxicokinetics using a digital twin method. To this end, we adopted the following technical solution, which consists of five steps: (1) constructing a database containing data on different heavy metal species and zebrafish at various stages based on databases and research literature; (2) preprocessing the data to meet the standards for modeling; (3) constructing an energy metabolism and behavior coupling model based on zebrafish physiological characteristics and behavioral patterns to achieve a dynamic description of energy consumption under different activity states; (4) constructing a toxicokinetics-toxicodynamic model to establish the mechanism between toxic response and behavioral changes, achieving coupling between heavy metal kinetics and zebrafish physiological behavior; (5) developing a digital twin model of zebrafish heavy metal toxicity for toxicological research and constructing a visual simulation interface to achieve dynamic linkage between model calculation and three-dimensional animation. The specific method steps are as follows: I. Dataset Construction In this implementation, a foundational dataset for digital twin modeling is first constructed. The data primarily originates from publicly available toxicity databases and academic literature, including the ECOTOX database and relevant research literature indexed in Web of Science. The study uses the ECOTOX Knowledgebase to screen for toxicity data of zebrafish exposed to heavy metals (such as copper, cadmium, lead, and zinc), including exposure concentration, duration, and effects (such as survival rate). Simultaneously, a systematic search of relevant literature is conducted in Web of Science using keywords (such as "zebrafish" and "heavy metal toxicity") to supplement basic physiological parameters of zebrafish (such as body weight and metabolic rate). The search results are then filtered to remove records with excessive blanks or invalid entries, ensuring data quality.

[0057] During the data screening process, the following criteria were set: (1) The experimental subjects must be zebrafish (Danio rerio), including embryonic, juvenile and adult stages; (2) Exposure concentration (Cw) and exposure time (t) must be provided; (3) The toxicity endpoint must be a quantifiable indicator, such as LC50 or survival rate; (4) Duplicate data and data records with serious information loss were removed. In terms of feature selection, this invention constructs a dataset containing four types of variables: (1) Physiological parameters, including body weight (W), standard metabolic rate (SMR), maximum metabolic rate (MMR), and oxygen consumption rate (VO2); (2) Behavioral parameters, including swimming speed (v), cruising proportion (P1), sprint proportion (P2), resting proportion (P3), and circadian rhythm parameters (light time Tlight, darkness time Tdark); (3) Toxicity endpoints, including median lethal concentration (LC50), body concentration (Cint), and abnormal behavior indicators; In this embodiment, heavy metals such as copper, cadmium, lead, and zinc were selected to ensure the consistency of data sources and the comparability of model parameters. The final dataset, after screening, contains multiple sets of data under different conditions, forming a standardized data foundation for modeling.

[0058] II. Data Preprocessing

[0059] To ensure the data meets the model's computational requirements, the constructed dataset needs systematic preprocessing. First, data integrity is checked, and data with missing values ​​exceeding a preset threshold are removed. Second, numerical data undergoes unified processing: interval data is replaced with the mean, and missing values ​​are imputed using the average of similar features; categorical data is digitized using encoding methods to meet the model's input requirements.

[0060] Furthermore, the units of data from different sources were standardized. For example, heavy metal concentrations were standardized to μmol / L, time to hours (h), and swimming speed to cm / s to eliminate the influence of dimensional differences on the model. For experimental data involving time variations, such as in vivo concentration curves over time, linear interpolation was used to complete the data to meet the requirements of solving continuous functions. Finally, all numerical variables were standardized by normalizing the data using the mean and standard deviation to avoid bias in the model calculation results caused by differences in the scale of different variables.

[0061] After the above processing, the original data is transformed into a standard dataset that can be used for numerical computation and simulation modeling.

[0062] III. Construction of Energy Metabolism-Behavior Coupling Model and Toxicokinetic Model

[0063] Based on the physiological characteristics and behavioral patterns of zebrafish, a coupling model between energy metabolism and behavior was constructed. First, based on existing research, the daily feed intake of zebrafish was set at 5% of their body weight, with a feed energy value of 20 kJ / g and an absorption efficiency of 80%, to calculate the daily energy input. Second, based on the behavioral characteristics of zebrafish in their natural state, their behavior was divided into three basic states: cruising, sprinting, and stillness, each assigned a different time proportion. Cruising behavior is dominant, sprinting behavior is short-duration high-intensity activity, and stillness corresponds to a low metabolic state.

[0064] In metabolic modeling, the resting state corresponds to the standard metabolic rate (SMR), the cruising state is set to 2 to 3 times the SMR, and the sprint state corresponds to the maximum metabolic rate (MMR), thus achieving differentiated expression of energy consumption under different activity states. Simultaneously, a circadian rhythm regulation mechanism is introduced, dividing the 24 hours into light and dark periods, with a significant reduction in metabolic levels and activity ratio during the dark period. In practice, the behavioral ratio can be dynamically adjusted using a time function, enabling the model to simulate real physiological rhythm changes. Finally, by calculating energy input and energy output at different time points, the energy state of zebrafish at any given time point is obtained, providing input parameters for subsequent toxicity process modeling.

[0065] A one-compartment toxicokinetic model was used to describe the absorption and excretion of heavy metals in zebrafish. By establishing the dynamic relationship between in vivo and environmental concentrations, ordinary differential equations were used to characterize the accumulation process of heavy metals in the organism.

[0066] In the model implementation process, numerical methods are used to solve the differential equations. During the numerical implementation, a continuous curve of the in vivo concentration changing over time is obtained through iterative calculations.

[0067] IV. Digital Twin Model Development and Visualization

[0068] Based on the above model, a zebrafish digital twin system is constructed to achieve dynamic coupling between the virtual model and the computational model. First, a virtual zebrafish model is built using 3D modeling software (such as Blender), and skeletal rigging and swimming animation design are completed to give the model basic movement capabilities. Then, the behavioral parameters calculated from the model (such as swimming speed and activity state) are mapped to the 3D animation parameters, thereby enabling the virtual model to change in real time according to the calculation results.

[0069] During system implementation, an interface is used to facilitate interaction between data and the 3D model, enabling changes in in vivo concentrations to drive behavioral changes, which are then visually represented. This allows for dynamic updates of the virtual zebrafish's behavior, including changes in swimming speed and abnormal behavioral patterns. Simultaneously, an adjustable interactive interface is constructed, allowing users to input parameters such as the type of heavy metal, body weight, metabolic rate, and circadian rhythm. The system automatically performs model calculations and outputs results including changes in in vivo concentrations, energy changes, and behavioral responses. The simulation results are displayed graphically, making the system suitable for dynamic analysis and process understanding in toxicological research.

[0070] In summary, this invention constructs a zebrafish digital twin model, achieves parameter unification through data preprocessing, and establishes an energy metabolism and behavior coupling model to correlate physiological states with behavioral changes. Furthermore, it combines a one-compartment toxicokinetics model to simulate the zebrafish process under heavy metal conditions. Simultaneously, it integrates the above model with a three-dimensional virtual simulation and visualization interface to realize a digital twin system with visible process and interactive results, thus forming a toxicological research method that integrates modeling, simulation, and display.

[0071] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for constructing a digital twin model of zebrafish toxicity effects, characterized in that, Includes the following steps: S1: Multi-source data acquisition and modeling database construction: Collect multi-dimensional toxicological data of zebrafish through public toxicity databases and academic literature, and construct a dedicated basic database for digital twins; S2: Data standardization preprocessing: Perform preprocessing on the raw data in the database to output a standard dataset suitable for modeling operations; S3: Construction of a coupling model of energy metabolism and behavior: Based on the physiological and behavioral characteristics of zebrafish, an energy balance equation is introduced to establish an energy balance, which is used to characterize the dynamic energy balance of zebrafish under different behavioral states; S4: Construction of Toxicity-Toxicity Kinetic Model: Used to characterize the dynamic changes of pollutants in zebrafish and to establish a feedback mechanism between toxic effects and behavioral responses; S5: Digital Twin Simulation Model Development and Visual Interaction Setup: Construct a three-dimensional virtual entity model of a zebrafish, establish a mapping relationship between numerical calculation parameters and three-dimensional animation driving parameters, and develop a visual interactive interface that allows for customizable input conditions and real-time output of simulation results.

2. The method for constructing a digital twin model of zebrafish toxicity effects according to claim 1, characterized in that, The data collected by S1 includes: zebrafish physiological parameters, including body weight (W), body length (L), standard metabolic rate (SMR), maximum metabolic rate (MMR), oxygen consumption rate, and energy conversion efficiency. Behavioral characteristics data include: swimming speed v, behavioral time allocation ratio, and diurnal rhythm parameters; Environmental exposure parameters include: heavy metal concentration Cw, exposure time t, and water temperature T; Toxicity endpoint data include: median lethal concentration (LC50), survival rate, and behavioral abnormality indicators; The publicly available toxicity database is the ECOTOX database, and the academic literature data source is the Web of Science.

3. The method for constructing a digital twin model of zebrafish toxicity effects according to claim 2, characterized in that, The preprocessing methods for S2 include: missing value cleaning, numerical imputation, category coding, and unit unification and standardization; the specific steps are as follows: S21: Handling missing values: Remove data records with a missing ratio exceeding a preset threshold; S22: Numerical data is processed uniformly; for interval data, the mean is used as a substitute, and missing values ​​are filled with the average value of similar features; for categorical data, a coding method is used for digital processing. S23: Standardize environmental factors and physiological parameters to avoid the influence of different dimensions on the model. The standardization formula is as follows: in, It represents the data sequence number. i It is a positive integer. It is the first i 1 eigenvalue, It is the first i The average of the features, It is the first i The standard deviation (SD) of each feature Then it is the first i The eigenvalues ​​after normalization.

4. The method for constructing a digital twin model of zebrafish toxicity effects according to claim 3, characterized in that, In S3, the construction of the energy metabolism and behavior coupling model includes: Zebrafish behavior is divided into at least two behavioral states, and a corresponding metabolic rate is set for each behavioral state. Set energy intake parameters and calculate energy input. Calculate energy consumption based on the time allocation and corresponding metabolic rate of each behavioral state; The real-time energy state of zebrafish is determined based on the difference between energy input and energy consumption.

5. The method for constructing a digital twin model of zebrafish toxicity effects according to claim 4, characterized in that, Behavioral states include cruising, sprinting, and stationary states; The metabolic rate corresponding to the cruising state is 2-3 times the standard metabolic rate (SMR), the metabolic rate corresponding to the sprint state is the maximum metabolic rate (MMR), and the metabolic rate corresponding to the resting state is the standard metabolic rate (SMR). By introducing a circadian rhythm mechanism, the 24-hour period is divided into a light period and a dark period, and the metabolic level is reduced during the dark period.

6. The method for constructing a digital twin model of zebrafish toxicity effects according to claim 5, characterized in that, The S4 toxicokinetics-toxicodynamics model uses a one-compartment model to describe the dynamic changes of pollutants in zebrafish: in, This refers to the concentration in the body. For environmental concentration, The absorption rate constant is The excretion rate constant is given; the steady-state enrichment concentration in the body is: Through parameters and Calculate bioaccumulation factors BCF : This refers to the concentration in the body. For environmental concentration.

7. The method for constructing a digital twin model of zebrafish toxicity effects according to claim 6, characterized in that, Calculate the real-time energy state coefficient of zebrafish in S4: in, Represents the energy state coefficient, real-time energy input. With real-time energy consumption ,when A value >0 indicates that the body is in a state of energy surplus. A value less than 0 indicates that the body is in a state of energy deficiency. Absorption rate constant and discharge rate constant Correct it as follows: in, and These are the basic absorption rate constant and the basic excretion rate constant, respectively; α and β It is the energy regulation coefficient, used to characterize the degree of influence of energy state on the pollutant absorption and emission capacity.

8. The method for constructing a digital twin model of zebrafish toxicity effects according to claim 7, characterized in that, The feedback mechanism between the toxic effects and behavioral responses in S4 includes: When the concentration in the body When the preset toxicity response threshold is reached or exceeded, the zebrafish is determined to have entered a toxicity response state. In the state of toxic response, the time allocation ratio of each behavioral state is adjusted to reduce the proportion of high-energy-consuming behaviors and increase the proportion of low-energy-consuming behaviors. The energy consumption is recalculated based on the adjusted time allocation ratio, and the energy state parameters in the digital twin model are updated. The method for adjusting the time allocation ratio for activities is as follows: in, The original parade time, The original sprint time, For the original rest time, The original parade time, The original sprint time, For the original rest time, T Total action time, λ and μ These represent the inhibition coefficients for cruising behavior and sprinting behavior, respectively. The energy consumption per unit time is recalculated based on the updated behavior time, and the expression is as follows: in, SMR Indicates standard metabolic rate. MMR Indicates the maximum metabolic rate. This indicates the energy expenditure of specific dynamic processes after ingestion.

9. The method for constructing a digital twin model of zebrafish toxicity effects according to claim 8, characterized in that, In S5, the construction of the zebrafish digital twin simulation model includes: A three-dimensional virtual model of a zebrafish is constructed, and a mapping relationship is established between the model's calculation parameters and the virtual model's motion parameters, so that the virtual model's behavior response changes dynamically with the model's calculation results. Construct an interactive interface with adjustable parameters, where the adjustable parameters include at least one of pollutant type, exposure concentration, and exposure time; The behavioral responses of the virtual model include at least one of the following: changes in swimming speed, changes in tail wag frequency, and switching of behavioral states; The interactive interface is also used to output at least one of the following in real time: in vivo concentration change data, energy change data, and behavioral response results.

10. A digital twin model construction system for zebrafish toxicity effects, characterized in that, A method for constructing a digital twin model of zebrafish toxicity effects according to any one of claims 1-9 includes a database module, a data preprocessing module, an energy metabolism coupling calculation module, a toxicokinetics feedback calculation module, and a three-dimensional digital twin visualization simulation module connected in sequence. The database module collects multi-dimensional toxicological data of zebrafish from public toxicity databases and academic literature to build a dedicated basic database for digital twins. The data preprocessing module performs preprocessing on the raw data in the database and outputs a standard dataset adapted for modeling operations. The energy metabolism coupling calculation module, based on the physiological and behavioral characteristics of zebrafish, introduces an energy balance equation to establish energy balance, which is used to characterize the dynamic energy balance of zebrafish under different behavioral states. The toxicokinetics feedback calculation module is used to characterize the dynamic changes of pollutants in zebrafish and to establish a feedback mechanism between toxic effects and behavioral responses. 3D Digital Twin Visualization Simulation Module: Constructs a 3D virtual entity model of a zebrafish, establishes a mapping relationship between numerical calculation parameters and 3D animation driving parameters, and develops a visual interactive interface that allows for customizable input conditions and real-time output of simulation results.

Images