Method for risk assessment of antibiotics in environmental media
Patent Information
- Application Number
- PCT/CN2026/104123
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-11-14
- Filing Date
- 2026-06-17
- Publication Date
- 2026-10-01
Smart Images
Figure CN2026104123_01102026_PF_FP_ABST
Abstract
Description
A method for assessing the risk of antibiotics in environmental media Technical Field
[0001] This invention relates to the field of antibiotic risk assessment technology, and in particular to a method for assessing the risk of antibiotics in environmental media. Background Technology
[0002] Currently, antibiotic overuse has become a global problem, resulting in high levels of antibiotic residues in the environment. Many countries and regions have shortcomings in antibiotic use and management, exacerbating the problem of antibiotic pollution in the environment. Antibiotic residues not only pose a potential threat to ecosystems but may also lead to increased bacterial resistance through environmental transmission, thereby threatening human health.
[0003] Environmental benchmark studies for antibiotics currently rely primarily on traditional toxicological methods. While these methods are valuable for assessing the toxicity of high concentrations of antibiotics, they have significant limitations in assessing the risk of antibiotic resistance at low concentrations in the environment. Antibiotic contamination levels in surface water typically range from nanograms per liter to micrograms per liter, far lower than the values derived from traditional toxicological methods. However, even at such low concentrations, clinical and epidemiological studies have found the risk of antibiotic resistance.
[0004] Furthermore, many countries and regions have long lacked regulatory documents regarding antibiotic use and pollution, resulting in a technological gap in antibiotic environmental risk assessment. Current national drinking water standards also lack antibiotic testing standards, meaning that even if water quality fully meets national standards, the potential risks of antibiotics cannot be eliminated. Therefore, developing scientific and effective methods for antibiotic environmental risk assessment is crucial for management departments to conduct risk control and decision-making, and will help curb the global spread of antibiotic resistance. Summary of the Invention
[0005] The purpose of this invention is to provide a method for assessing the risk of antibiotics in environmental media. By combining literature data, metagenomic sequencing, and machine learning technology, the method can accurately assess the environmental risks of antibiotics and their resistance genes, providing strong support for antibiotic pollution control and public health protection, and helping to curb the spread of antibiotic resistance risks.
[0006] To achieve the above objectives, the present invention provides a method for assessing the risk of antibiotics in environmental media, comprising the following steps:
[0007] Step S1: Obtain data on antibiotics and resistance genes in environmental media through literature retrieval or metagenomic sequencing, and identify resistance genes and their hosts by combining bioinformatics analysis.
[0008] Step S2: Using site type, geographical location, environmental physicochemical properties, antibiotic concentration, co-selected pollutant concentration and microbial community structure as input variables and resistance gene abundance as output variable, a machine learning algorithm is used to construct a predictive model of resistance gene abundance in the environmental medium.
[0009] Step S3: Input the abundance of resistance genes as the input variable into the dose-response assessment model and output the infection probability;
[0010] Step S4: Using the antibiotic concentration at an infection probability of 10% as the concentration threshold to protect 90% of organisms from being affected, the abundance of resistance genes corresponding to an infection probability of 10% is determined through the cumulative infection probability curve and used as the environmental ecological threshold concentration of antibiotics.
[0011] Step S5: Based on the antibiotic environmental ecological threshold concentration obtained in step S4, assess the ecological risk of antibiotics in the environmental medium using the safety threshold method.
[0012] Preferably, in step S1, the environmental medium includes surface water and soil.
[0013] Preferably, in step S1, the literature search uses appropriate search terms, including Chinese search terms and English search terms.
[0014] Preferably, in step S1, metagenomic sequencing includes sample preparation, library construction, sequencing, and data analysis.
[0015] Preferably, in step S1, resistance genes are screened based on environmental abundance, distribution breadth, migration and health association.
[0016] Preferably, in step S2, the machine learning algorithm includes at least one of support vector machine, random forest, multiple linear regression, BP neural network, or partial least squares algorithm.
[0017] Preferably, in step S2, among the input variables, the location type and geographical location are categorical variables, and the rest are numerical variables.
[0018] Preferably, in step S3, the dose-response assessment model is one of the following models:
[0019] Weibull distribution model, exponential distribution model, log-normal distribution model, logistic distribution model, or Poisson distribution model.
[0020] Preferably, in step S5, the safety threshold method is calculated... Risk assessment should be conducted on the value:
[0021] ;
[0022] in, This indicates a safety margin based on a 10% probability of infection. This refers to the ecological threshold concentration of antibiotics. This represents the antibiotic concentration corresponding to a 90% probability of infection. When the value is greater than 1, it is considered that there is no ecological risk of antibiotics in the area to be evaluated. When the value is ≤1, the area to be evaluated is considered to have an antibiotic ecological risk.
[0023] Therefore, the present invention employs the above-mentioned method for assessing the risk of antibiotics in environmental media, and the beneficial technical effects are as follows:
[0024] (1) This invention focuses on types of antibiotics that are frequently found in the environment and pose significant risks. It can accurately identify and assess the risks of these antibiotics, avoid ineffective assessments of irrelevant antibiotics, and improve the efficiency and accuracy of risk assessment.
[0025] (2) In traditional methods, antibiotics have low biotoxicity, and the toxicity endpoint is difficult to define, leading to a complex and inaccurate assessment process. This invention solves this problem of traditional methods by directly addressing the issue from the perspective of microbial risk assessment, without considering the toxicity endpoint.
[0026] (3) This invention not only focuses on the concentration of antibiotics, but also considers the role of resistance genes, which can comprehensively assess the potential risks of antibiotics and their resistance genes to the environment and human health, and provide more comprehensive risk assessment results.
[0027] (4) This invention studies from the perspective of microbial risk assessment, focusing more on resistance genes colonized in pathogens, which can more accurately assess the risk of antibiotic resistance and provide a scientific basis for formulating effective risk control measures. Attached Figure Description
[0028] Figure 1 is a flowchart of a method for antibiotic risk assessment in environmental media according to the present invention; Detailed Implementation
[0029] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.
[0030] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning as understood by one of ordinary skill in the art to which this invention pertains.
[0031] Example 1
[0032] As shown in Figure 1, this embodiment takes tetracycline in the surface water environment as an example to specifically illustrate the implementation of the present invention.
[0033] Step S1: Data acquisition and resistance gene identification.
[0034] 1. Sample Collection and Sequencing: Surface water samples are collected, and metagenomic sequencing technology is used to obtain data on antibiotics and resistance genes in the environmental media. This process includes sample preparation, library construction, sequencing, and bioinformatics data analysis.
[0035] 2. Supplementing Literature Data: To expand the dataset, relevant data was obtained through literature retrieval. The search formula is as follows:
[0036] Chinese search query: ((“antibiotics” OR “resistance genes” OR “ARGs”) AND (“environmental media” OR “soil” OR “water bodies” OR “rivers” OR “lakes” OR “sediments” OR “wastewater” OR “surface water”));
[0037] English search formula: (("antibiotic" OR "resistance gene" OR "ARGs" OR "antibiotic resistance gene") AND ("environmental medium" OR "soil" OR "river" OR "lake" OR "water" OR "sediment" OR "sludge" OR "wastewater" OR "surface water")).
[0038] 3. Screening of resistance genes: Based on metagenomic data and literature data, and according to the criteria of environmental abundance, distribution, migration and health association of resistance genes, key resistance genes related to tetracycline were screened for subsequent analysis, and their potential host microorganisms were identified.
[0039] Step S2: Construct a resistance gene abundance prediction model.
[0040] 1. Variable setting: Environmental factors such as temperature, pH, total phosphorus (TP), total nitrogen (TN), tetracycline concentration, abundance of class I integrons, and pathogen proportion are used as input variables (features), and the abundance of key tetracycline resistance genes determined in step S1 is used as output variables (labels).
[0041] 2. Model Training and Comparison: Predictive models were constructed using various machine learning algorithms, and their performance was compared. Machine learning algorithms included the MLR multiple linear regression model and a BP neural network (500 iterations, error threshold 10). -4 Learning rate 0.01), partial least squares (PLS) algorithm, support vector machine (penalty factor 2.0, radial basis function parameter 0.167) or random forest (number of decision trees 40, minimum number of leaves 3), etc.
[0042] The performance of each model on the test set is shown in Table 1:
[0043] Table 1. Performance Comparison of Resistance Gene Abundance Prediction Models
[0044]
[0045] 3. Model Selection: Based on Table 1, the BP neural network model has the highest R-value in this embodiment. 2 The value (0.56) indicates that it has the strongest ability to explain data variation, and therefore it was selected as the final resistance gene abundance prediction model.
[0046] Step S3: Dose-response relationship assessment.
[0047] 1. Model Fitting: The collected "resistance gene abundance - population infection rate" data pairs are input into various classic dose-response models for fitting.
[0048] 2. Model comparison and selection: We selected the Weibull distribution, exponential distribution, log-normal distribution, logistic distribution, and Poisson distribution.
[0049] The goodness of fit of each model is shown in Table 2:
[0050] Table 2 Comparison of goodness of fit of dose-response models
[0051]
[0052] 3. Model Determination: R² of the logistic distribution model 2 The highest value (0.9714) and the lowest RMSE (0.0022) showed the best fit, and were therefore selected as the dose-response relationship model for this risk assessment.
[0053] Step S4: Determine the environmental ecological threshold for antibiotics.
[0054] 1. Determine the abundance threshold of resistance genes: Using the logistic distribution model determined in step S3, plot the cumulative infection probability curve and determine that when the infection probability is 10%, the corresponding ecological threshold for the abundance of resistance genes is 3.79 × 10⁻⁶. 5 .
[0055] 2. Back-calculation of antibiotic concentration threshold: Substituting the above-mentioned resistance gene abundance threshold into the BP neural network prediction model constructed in step S2, under the set environmental conditions (temperature 25℃, pH=7.0, TP=0.2mg / L, TN=1.0mg / L, pathogen proportion 1.5%, class I integrin abundance 0.02), the ecological threshold concentration of tetracycline was back-calculated to be 351.98ng / L. This concentration is the baseline concentration of tetracycline in surface water environments that protects 90% of the population from infection caused by this resistance gene.
[0056] Step S5: Environmental risk characterization.
[0057] The final risk characterization is performed using the safety threshold method. The calculation formula is as follows:
[0058] ;
[0059] in, This indicates a safety margin based on a 10% probability of infection. This refers to the ecological threshold concentration of antibiotics. This represents the antibiotic concentration corresponding to a 90% probability of infection.
[0060] when When the value is greater than 1, it is considered that there is no ecological risk of antibiotics in the area to be evaluated. When the value is ≤1, the area to be evaluated is considered to have an antibiotic ecological risk.
[0061] It is worth noting that all contents not described in detail in this invention are existing technologies and are well known to those skilled in the art.
[0062] Therefore, the present invention adopts the above-mentioned method for antibiotic risk assessment in environmental media, which is highly targeted, does not require consideration of toxicity endpoints, comprehensively assesses antibiotic concentration and the role of resistance genes, provides scientific basis from the perspective of microbial risk assessment, and helps to effectively manage antibiotic risks.
[0063] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.
Claims
1. A method for assessing the risk of antibiotics in environmental media, characterized in that, Includes the following steps: Step S1: Obtain data on antibiotics and resistance genes in environmental media through literature retrieval or metagenomic sequencing, and identify resistance genes and their hosts by combining bioinformatics analysis. Step S2: Using site type, geographical location, environmental physicochemical properties, antibiotic concentration, co-selected pollutant concentration and microbial community structure as input variables and resistance gene abundance as output variable, a machine learning algorithm is used to construct a predictive model of resistance gene abundance in the environmental medium. Step S3: Input the abundance of resistance genes as the input variable into the dose-response assessment model and output the infection probability; Step S4: Using the antibiotic concentration at an infection probability of 10% as the concentration threshold to protect 90% of organisms from being affected, the abundance of resistance genes corresponding to an infection probability of 10% is determined through the cumulative infection probability curve and used as the environmental ecological threshold concentration of antibiotics. Step S5: Based on the antibiotic environmental ecological threshold concentration obtained in step S4, assess the ecological risk of antibiotics in the environmental medium using the safety threshold method.
2. The method for assessing the risk of antibiotics in environmental media according to claim 1, characterized in that, In step S1, the environmental media include surface water and soil.
3. The method for assessing the risk of antibiotics in environmental media according to claim 1, characterized in that, In step S1, the literature search uses appropriate search terms, including Chinese search terms and English search terms.
4. The method for assessing the risk of antibiotics in environmental media according to claim 1, characterized in that, In step S1, metagenomic sequencing includes sample preparation, library construction, sequencing, and data analysis.
5. The method for assessing the risk of antibiotics in environmental media according to claim 1, characterized in that, In step S1, resistance genes are screened based on environmental abundance, distribution breadth, migration, and health association.
6. The method for assessing the risk of antibiotics in environmental media according to claim 1, characterized in that, In step S2, the machine learning algorithm includes at least one of support vector machine, random forest, multiple linear regression, BP neural network or partial least squares algorithm.
7. The method for assessing the risk of antibiotics in environmental media according to claim 1, characterized in that, In step S2, among the input variables, location type and geographical location are categorical variables, while the rest are numerical variables.
8. The method for assessing the risk of antibiotics in environmental media according to claim 1, characterized in that, In step S3, the dose-response assessment model is one of the following models: Weibull distribution model, exponential distribution model, log-normal distribution model, logistic distribution model, or Poisson distribution model.
9. The method for assessing the risk of antibiotics in environmental media according to claim 1, characterized in that, In step S5, the safety threshold method is calculated... Risk assessment should be conducted on the value: ； in, This indicates a safety margin based on a 10% probability of infection. This refers to the ecological threshold concentration of antibiotics. This represents the antibiotic concentration corresponding to a 90% probability of infection. When the value is greater than 1, it is considered that there is no ecological risk of antibiotics in the area to be evaluated. When the value is ≤1, the area to be evaluated is considered to have an antibiotic ecological risk.