Using Machine Learning to Predict Adverse Effects of Metallic Nanomaterials to Various Aquatic Organisms

Predicting estrogen receptor agonists from plastic additives across various aquatic-related species using machine learning and AlphaFold2

The absence of effective public databases greatly limits high-throughput prediction of hormonal effects mediated by nuclear receptors in aquatic organisms. In this study, we developed novel strategies for multi-species screening of estrogen receptor (ER) agonists in plastic additives using AlphaFold2. Firstly, Deep Forest (DF), artificial neural network (ANN) and conventional machine learning (ML) models were utilized to screen ERα agonists. The DF models using RDKit.Chem.Descriptors and MorganFingerprint achieved a sensitivity = 0.96, specificity > 0.99, and an F1 score > 0.95, identifying 42 plastic additives as ERα agonists. Subsequently, ERα structures for Danio rerio (Dr), Oryzias melastigma (Om), Delphinus delphis (Dd), Physeter catodon (Pc), Mytilus edulis (Me), Xenopus tropicalis (Xt), Nipponia nippon (Nn), and Aptenodytes forsteri (Af) were constructed using AlphaFold2. Except for Me ERα, most species shared two common key amino acid residues responsible for ERα activity: arginine 85 and glutamic acid 44 (aligned serial numbers in the LBD). However, aquatic-related species exhibited other three additional key residues: glycine 212, leucine 216 and phenylalanine 95 (aligned serial numbers in the LBD). The number of compounds with docking energy < -9 kcal/mol for Dr, Om, Dd, Pc, Me, Xt, Nn, and Af were 4, 8, 4, 12, 10, 13, 7, and 9, respectively. The docking energy of estrone in all species was < -9 kcal/mol, while that of bisphenol P varied greatly among different species. The combined application of ML and AlphaFold enables high-throughput evaluation of the ecotoxicity posed by emerging pollutants across multiple aquatic-related species.

Using machine learning models to predict the dose-effect curve of municipal wastewater for zebrafish embryo toxicity

Municipal wastewater substantially contributes to aquatic ecological risks. Assessing the toxicity of municipal wastewater through dose-effect curves is challenging owing to the time-consuming, labor-intensive, and costly nature of biological assays. This study developed machine learning models to predict wastewater dose-effect curves for zebrafish embryos. The influent and effluent samples from 176 wastewater treatment plants in China were analyzed to collect water quality data, including information on seven chemical parameters and the toxic effects on zebrafish embryos at eight relative enrichment factors (REFs) of wastewater. Using Spearman's rank correlation coefficient and the max-relevance and min-redundancy algorithm, the parameters of ammonium nitrogen content and toxic effect values at REFs of 2 and 25 (REF2 and REF25), were identified as crucial input features from 15 variables. Decision tree, random forest, and gradient-boosted decision tree (GBDT) models were developed. Among these, GBDT exhibited the best performance, with an average R2 value of 0.91 and an average mean absolute percentage error (MAPE) of 27.91 %. Integrating the dose-effect curve pattern into the machine learning model considerably optimized the GBDT model, reaching a minimum MAPE of 14.74 %. The developed model can accurately determine the dose-effect curves of actual wastewater, reducing at least 75 % of the experimental workload. These findings provide a valuable tool for assessing zebrafish embryo toxicity in municipal wastewater management. This study indicates that combining environmental expertise and machine learning models allows for a scientific assessment of the potential toxic risks in wastewater, providing new perspectives and approaches for environmental policy development.

Exposure experiments and machine learning revealed that personal care products can significantly increase transdermal exposure of SVOCs from the environment

We investigated the impacts of personal care products (PCPs) on dermal exposure to semi-volatile organic compounds (SVOCs), including phthalates, organophosphate esters, polycyclic aromatic hydrocarbons (PAHs), ultraviolet filters, and p-phenylenediamines, through an experiment from volunteers, explored the impact mechanisms of PCP ingredients on dermal exposure, and predicted the PCP effects on SVOC concentrations in human serum using machine learning. After applying PCPs, namely lotion, baby oil, sunscreen, and blemish balm, the dermal adsorption of SVOCs increased significantly by 1.63 ± 0.62, 1.97 ± 0.73, 1.91 ± 0.48, and 2.03 ± 0.59 times, respectively, probably due to the absorption effects of PCP ingredients. Ingredient tocopherol can increase dermal adsorption of SVOCs by 2.59 ± 1.60 times. PCPs can either increase or decrease the SVOC transdermal exposure risks, depending on the properties of their ingredients. Blemish balm caused the highest hazard quotient for certain SVOCs, while tris(2-chloroethyl) phosphate (TCEP) exhibited the highest hazard quotient. We predicted the SVOC concentrations in serum before and after applying PCPs based on the PCP-increased skin permeation doses and machine learning. PCPs can significantly increase the serum concentrations of PAHs with 2-3 rings and TCEP. This study first revealed that PCPs can significantly increase the dermal exposure of SVOCs from the surroundings, resulting in potentially higher health risks.

Design and performance analysis of multi-enzyme activity-doped nanozymes assisted by machine learning

Traditional design approaches for nanozymes typically rely on empirical methods and trial-and-error, which hampers systematic optimization of their structure and performance, thus limiting the efficiency of developing innovative nanozymes. This study leverages machine learning techniques supported by high-throughput computations to effectively design nanozymes with multi-enzyme activities and to elucidate their reaction mechanisms. Additionally, it investigates the impact of dopants on the microphysical properties of nanozymes. We constructed a machine learning prediction framework tailored for dopant nanozymes exhibiting catalytic activities like to oxidase (OXD) and peroxidase (POD). This framework was used to evaluate key catalytic performance parameters, such as formation energy, density of states (DOS), and adsorption energy, through density functional theory (DFT) calculations. Various machine learning models were employed to predict the effects of different doping element ratios on the catalytic activity and stability of nanozymes. The results indicate that the combination of machine learning with high-throughput computations significantly accelerates the design and optimization of dopant nanozymes, providing an efficient strategy to address the complexities of nanozyme design. This approach not only boosts the efficiency and capability for innovation in material design but also provides a novel theoretical analytical avenue for the development of new functional materials.

Addressing data gaps in deriving aquatic life ambient water quality criteria for contaminants of emerging concern: Challenges and the potential of in silico methods

The international community is becoming increasingly aware of the threats posed by contaminants of emerging concern (CECs) for ecological security. Aquatic life ambient water quality criteria (WQC) are essential for the formulation of risk prevention and control strategies for pollutants by regulatory agencies. Accordingly, we systematically evaluated the current status of WQC development for typical CECs through literature review. The results revealed substantial disparities in the WQC for the same chemical, with the coefficients of variation for all CECs exceeding 0.3. The reliance on low-quality data, high-uncertainty derivation methods, and limited species diversity highlights a substantial data gap. Newly developed in silico methods, with potential to predict the toxicity of untested chemicals, species, and conditions, were classified and integrated into a traditional WQC derivation framework to address the data gap for CECs. However, several challenges remain before such methods can achieve widespread acceptance. These include unstable model performance, the inability to predict chronic toxicity, undefined model applicability, difficulties in specifying toxicity effects and predicting toxicity for certain key species. Future research should prioritize: 1) improving model accuracy by developing specialized models trained with relevant, chemical-specific data or integrating chemical-related features into interspecies models; 2) enhancing species generalizability by developing multispecies models; 3) facilitating the derivation of environmentally relevant WQC by incorporating condition-related features into models; and 4) improving the regulatory acceptability of in silico methods by evaluating the reliability of "black-box" models.

Common issues of data science on the eco-environmental risks of emerging contaminants

Data-driven approaches (e.g., machine learning) are increasingly used to replace or assist laboratory studies in the study of emerging contaminants (ECs). In the past ten years, an increasing number of models or approaches have been applied to ECs, and the datasets used are continuously enriched. However, there are large knowledge gaps between what we have found and the natural eco-environmental meaning. For most published reviews, the contents are organized by the types of ECs, but the common issues of data science, regardless of the type of pollutant, are not sufficiently addressed. To close or narrow the knowledge gaps, we highlight the following issues ignored in the field of data-driven EC research. Complicated biological and ecological data and ensemble models revealing mechanisms and spatiotemporal trends with strong causal relationships and without data leakage deserve more attention in the future. In addition, the matrix influence, trace concentration, and complex scenario have often been ignored in previous works. Therefore, an integrated research framework related to natural fields, ecological systems, and large-scale environmental problems, rather than relying solely on laboratory data-related analysis, is urgently needed. Beyond the current prediction purposes, data science can inspire the discovery of scientific questions, and mutual inspiration among data science, process and mechanism models, and laboratory and field research is a critical direction. Focusing on the above urgent and common issues related to data, frameworks, and purposes, regardless of the type of pollutant, data science is expected to achieve great advancements in addressing the eco-environmental risks of ECs.

Unraveling Microplastic Effects on Gut Microbiota across Various Animals Using Machine Learning

Microplastics, rapidly expanding and durable pollutant, have been shown to significantly impact gut microbiota across a spectrum of animal species. However, comprehensive analyses comparing microplastic effects on gut microbiota among these species are still limited, and the critical factors driving these effects remain to be clarified. To address these issues, we compiled 1352 gut microbiota samples from six animal categories, employing machine learning to conduct an in-depth meta-analysis. Our study revealed that mice, compared with other animals, not only exhibit a heightened susceptibility to the toxic effects of microplastics─evidenced by decreased gut microbiota diversity, increased Firmicutes/Bacteroidetes ratios, destabilized microbial networks, and disruption in the equilibrium of beneficial and harmful bacteria─but also possess limited potential to degrade microplastics, unlike earthworms and insects. Furthermore, machine learning models confirmed that exposure duration is the key factor driving changes induced by microplastics in gut microbiota. We also identified Lactobacillus, Helicobacter, and Pseudomonas as potential biomarkers for detecting microplastic toxicity in the animal gut. Overall, these findings provide valuable insights into the health risks and driving factors associated with microplastic exposure across multiple animal species.

Integrating machine learning and nano-QSAR models to predict the oxidative stress potential caused by single and mixed carbon nanomaterials in algal cells

In silico methods are increasingly important in predicting the ecotoxicity of engineered nanomaterials (ENMs), encompassing both individual and mixture toxicity predictions. It is widely recognized that ENMs trigger oxidative stress effects by generating intracellular reactive oxygen species (ROS), serving as a key mechanism in their cytotoxicity studies. However, existing in silico methods still face significant challenges in predicting the oxidative stress effects induced by ENMs. Herein, we utilized laboratory-derived toxicity data and machine learning methods to develop quantitative nanostructure-activity relationship (nano-QSAR) classification and regression models, aiming to predict the oxidative stress effects of five carbon nanomaterials (fullerene, graphene, graphene oxide, single-walled carbon nanotubes, and multi-walled carbon nanotubes) and their binary mixtures on Scenedesmus obliquus cells. We constructed five nano-QSAR classification models by combining zeta potential (ζP) with the C4.5 decision tree, support vector machine, artificial neural network, naive Bayes, and K-nearest neighbor algorithms. Moreover, we constructed three classification models by integrating the features including ζP, hydrodynamic diameter (DH), and specific surface area (SSA) with the logistic regression, random forest, and Adaboost algorithms. The Accuracy, Recall, Precision and harmonic mean of Precision and Recall (F1-score) values of these models were all higher than 0.600, indicating an excellent performance in distinguishing whether CNMs have the potential to generate ROS. In addition, using the ζP, DH, and SSA descriptors, we combined decision tree regression, random forest regression, gradient boosting, and the Adaboost algorithm, and successfully constructed four nano-QSAR regression models with applicable application domains (all training and testing data points lie within 95% confidence intervals), goodness-of-fit (Rtrain2 ≥ 0.850), and robustness (cross-validation R2 ≥ 0.650) as well as predictive power (Rtest2 ≥ 0.610). The method developed would establish a fundamental basis for more precise evaluations of ecological risks posed by these materials from a mechanistic standpoint.

A Comparative Review: Biological Safety and Sustainability of Metal Nanomaterials Without and with Machine Learning Assistance

In recent years, metal nanomaterials and nanoproducts have been developed intensively, and they are now widely applied across various sectors, including energy, aerospace, agriculture, industry, and biomedicine. However, nanomaterials have been identified as potentially toxic, with the toxicity of metal nanoparticles posing significant risks to both human health and the environment. Therefore, the toxicological risk assessment of metal nanomaterials is essential to identify and mitigate potential adverse effects. This review provides a comprehensive analysis of the safety and sustainability of metallic nanoparticles (such as Au NPs, Ag NPs, etc.) in key domains such as medicine, energy, and environmental protection. Using a dual-perspective analysis approach, it highlights the unique advantages of machine learning in data processing, predictive modeling, and optimization. At the same time, it underscores the importance of traditional methods, particularly their ability to offer greater interpretability and more intuitive results in specific contexts. Finally, a comparative analysis of traditional methods and machine learning techniques for detecting the toxicity of metal nanomaterials is presented, emphasizing the key challenges that need to be addressed in future research.

Toxicity prediction and risk assessment of per- and polyfluoroalkyl substances for threatened and endangered fishes

Per- and polyfluoroalkyl substances (PFASs) are severely polluted in aquatic environments and can harm aquatic organisms. Due to the limitation of conducting toxicity experiments directly on threatened and endangered (T&E) species, their toxicity data is scarce, hindering accurate risk assessments. The development of computational toxicology makes it possible to assess the risk of pollutants to T&E fishes. This study innovatively combined machine learning models, including random forest (RF), artificial neural network (ANN), and XGBoost, and the QSAR-ICE model to predict chronic developmental toxicity data of PFASs to T&E fishes. Among these, the XGBoost model exhibited superior performance, with R2 of 0.95 and 0.81 for the training and testing sets, respectively. Internal and external validation further confirmed that the XGBoost model is robust and reliable. Subsequently, it was used to predict chronic developmental toxicity data for seven priority PFASs to T&E fishes in the Yangtze River. Acipenseridae fishes (e.g., Acipenser dabryanus and Acipenser sinensis) showed high sensitivity to PFASs, possibly due to their unique lifestyle and physiological characteristics. Based on these data, the predicted no-effect concentration (PNEC) of individual PFASs was calculated, and the risk for T&E fishes in the Yangtze River was assessed. The results indicated that the risk of PFASs to T&E fishes is low (3.85 × 10-9∼8.20 × 10-4), with perfluorohexanoic acid (PFHxA) and perfluorooctanoic acid (PFOA) as the high-risk pollutants. The risk in the middle and lower reaches of the river is higher than in the upper reaches. This study provides a new approach for obtaining chronic toxicity data and conducting risk assessments for T&E species, advancing the protection of T&E species worldwide.

Prediction of developmental toxic effects of fine particulate matter (PM2.5) water-soluble components via machine learning through observation of PM2.5 from diverse urban areas

The global health implications of fine particulate matter (PM2.5) underscore the imperative need for research into its toxicity and chemical composition. In this study, zebrafish embryos exposed to the water-soluble components of PM2.5 from two cities (Harbin and Hangzhou) with differences in air quality, underwent microscopic examination to identify primary target organs. The Harbin PM2.5 induced dose-dependent organ malformation in zebrafish, indicating a higher level of toxicity than that of the Hangzhou sample. Harbin PM2.5 led to severe deformities such as pericardial edema and a high mortality rate, while the Hangzhou sample exhibited hepatotoxicity, causing delayed yolk sac absorption. The experimental determination of PM2.5 constituents was followed by the application of four algorithms for predictive toxicological assessment. The random forest algorithm correctly predicted each of the effect classes and showed the best performance, suggesting that zebrafish malformation rates were strongly correlated with water-soluble components of PM2.5. Feature selection identified the water-soluble ions F- and Cl- and metallic elements Al, K, Mn, and Be as potential key components affecting zebrafish development. This study provides new insights into the developmental toxicity of PM2.5 and offers a new approach for predicting and exploring the health effects of PM2.5.

Unlocking the potential of carbon dots in agriculture using data-driven approaches

The utilization of carbon dots (CDs) in agriculture to enhance plant growth has gained significant attention, but the data remains fractionated. Systematically integrating existing data is needed to identify the factors driving the interactions between CDs and plants and strategically guide future research. Articles reporting on CDs and their effects on plants were searched based on inclusion and exclusion criteria, resulting in the collection of 71 articles comprising a total of 2564 data points. The meta-analysis reveals that the soil and foliar application of red-emitting bio-derived CDs at a low concentration (<10 ppm) leads to the most beneficial effects on plant growth. Random forest and gradient boosting algorithms revealed that the size and dose of CDs were important factors in predicting plant responses across multiple aspects (CDs properties, plant properties, environmental factors, and experimental conditions). Specifically, smaller sizes are more favorable to growth indicators (GI) below 6 nm, nutrient and quality (NuQ) at 3-6 nm, photosynthesis (PSN) below 7 nm, and antioxidant responses (AR) below 5 nm. Overall, our analysis of existing data suggests that CDs applications can significantly improve plant responses (GI, NuQ, PSN, and AR) by 10-39 %. To unlock the full potential of CDs, customized synthesis techniques should be employed to meet the specific requirements of different crops and climate condition. For example, we recommend the synthesis of small CDs (<7 nm) with emission peak values falling within the range of 405-475 and 610-670 nm to enhance plant growth. The global prediction of plant responses to CDs application in future scenarios have shown significant improvements ranging from 17 to 58 %, suggesting that CDs have widespread applicability. This novel understanding of the impact of CDs on plant response provides valuable insights for optimizing the application of these nanomaterials in agriculture.

3MTox: A motif-level graph-based multi-view chemical language model for toxicity identification with deep interpretation

Toxicity identification plays a key role in maintaining human health, as it can alert humans to the potential hazards caused by long-term exposure to a wide variety of chemical compounds. Experimental methods for determining toxicity are time-consuming, and costly, while computational methods offer an alternative for the early identification of toxicity. For example, some classical ML and DL methods, which demonstrate excellent performance in toxicity prediction. However, these methods also have some defects, such as over-reliance on artificial features and easy overfitting, etc. Proposing novel models with superior prediction performance is still an urgent task. In this study, we propose a motifs-level graph-based multi-view pretraining language model, called 3MTox, for toxicity identification. The 3MTox model uses Bidirectional Encoder Representations from Transformers (BERT) as the backbone framework, and a motif graph as input. The results of extensive experiments showed that our 3MTox model achieved state-of-the-art performance on toxicity benchmark datasets and outperformed the baseline models considered. In addition, the interpretability of the model ensures that the it can quickly and accurately identify toxicity sites in a given molecule, thereby contributing to the determination of the status of toxicity and associated analyses. We think that the 3MTox model is among the most promising tools that are currently available for toxicity identification.

Understanding the phytotoxic effects of organic contaminants on rice through predictive modeling with molecular descriptors: A data-driven analysis

The widespread introduction of organic compounds into environments poses significant risks to ecosystems. Assessing the adverse effects of organic contaminants on crops is crucial for ensuring food safety. However, laboratory research is often time-consuming and costly, and machine learning (ML) methods can offer a viable solution to address these challenges. This study aimed at developing a ML model that incorporates chemical descriptors to predict the phytotoxicity of organic contaminants on rice. A dataset was compiled by gathering published experimental data on the phytotoxicity of 60 organic compounds, with a focus on morphological inhibition, photosynthesis perturbation, and oxidative stress. Four ML models (RF, SVM, GBM, ANN) were developed using chemical molecular descriptors (CMD) and the Molecular ACCess System (MACCS) keys. RF-MACCS model demonstrated the highest fitness, achieving an R2 value of 0.79 and an RMSE of 0.14. Feature importance analysis highlighted nAtom, HBA, logKow, and TPSA as the most influential CMDs in our model. Additionally, substructures containing oxygen atoms, carbonyl group and carbon chains with nitrogen and oxygen atoms were identified as significant factors associated with phytotoxicity. This data-driven study could aid in predicting the phytotoxicity of organic contaminants on crops and evaluating the potential risks of emerging contaminants in agroecosystems.

Perspectives on Advancing Multimodal Learning in Environmental Science and Engineering Studies

The environment faces increasing anthropogenic impacts, resulting in a rapid increase in environmental issues that undermine the natural capital essential for human wellbeing. These issues are complex and often influenced by various factors represented by data with different modalities. While machine learning (ML) provides data-driven tools for addressing the environmental issues, the current ML models in environmental science and engineering (ES&E) often neglect the utilization of multimodal data. With the advancement in deep learning, multimodal learning (MML) holds promise for comprehensive descriptions of the environmental issues by harnessing data from diverse modalities. This advancement has the potential to significantly elevate the accuracy and robustness of prediction models in ES&E studies, providing enhanced solutions for various environmental modeling tasks. This perspective summarizes MML methodologies and proposes potential applications of MML models in ES&E studies, including environmental quality assessment, prediction of chemical hazards, and optimization of pollution control techniques. Additionally, we discuss the challenges associated with implementing MML in ES&E and propose future research directions in this domain.

Nanosafety: a Perspective on Nano-Bio Interactions

Engineered nanomaterials offer numerous benefits to society ranging from environmental remediation to biomedical applications such as drug or vaccine delivery as well as clean and cost-effective energy production and storage, and the promise of a more sustainable way of life. However, as nanomaterials of increasing sophistication enter the market, close attention to potential adverse effects on human health and the environment is needed. Here a critical perspective on nanotoxicological research is provided; the authors argue that it is time to leverage the knowledge regarding the biological interactions of nanomaterials to achieve a more comprehensive understanding of the human health and environmental impacts of these materials. Moreover, it is posited that nanomaterials behave like biological entities and that they should be regulated as such.

Predicting the new psychoactive substance activity of antitussives and evaluating their ecotoxicity to fish

The misuse of antitussives preparations is a continuing problem in the world, and imply that they might have potential new psychoactive substances (NPS) activity. However, few study focus on their ecological toxicity towards fish. In the present study, the machine learning (ML) methods gcForest and random forest (RF) were employed to predict NPS activity in 30 antitussives. The potential toxic target, mode of action (MOA), acute toxicity and chronic toxicity to fish were further investigated. The results showed that both gcForest and RF achieved optimal performance when utilizing combined features of molecular fingerprint (MF) and molecular descriptor (MD), with area under the curve (AUC) = 0.99, accuracy >0.94 and f1 score > 0.94, and were applied to screen the NPS activity in antitussives. A total of 15 antitussives exhibited potential NPS activity, including frequently-used substances like codeine and dextromethorphan. The binding affinity of these antitussives with zebrafish dopamine transporter (zDAT) was high, and even surpassing that of some traditional narcotics and NPS. Some antitussives formed hydrogen bonds or salt bridges with aspartate (Asp) 95, tyrosine (Tyr) 171 of zDAT. For the ecotoxicity, the MOA of these 15 antitussives in fish was predicted as narcosis. The prenoxdiazin, pholcodine, codeine, dextromethorphan and dextrorphan exhibited very toxic/toxic to fish. It was necessary to pay close attention to the ecotoxicity of these antitussives. In this study, the integration of ML, molecular docking and ECOSAR approaches are powerful tools for understanding the toxicity profiles and ecological hazards posed by new pollutants.

Deciphering silver nanoparticles perturbation effects and risks for soil enzymes worldwide: Insights from machine learning and soil property integration

Globally extensive research into how silver nanoparticles (AgNPs) affect enzyme activity in soils with differing properties has been limited by cost-prohibitive sampling. In this study, customized machine learning (ML) was used to extract data patterns from complex research, with a hit rate of Random Forest > Multiple Imputation by Chained Equations > Decision Tree > K-Nearest Neighbors. Results showed that soil properties played a pivotal role in determining AgNPs' effect on soil enzymes, with the order being pH > organic matter (OM) > soil texture ≈ cation exchange capacity (CEC). Notably, soil enzyme activity was more sensitive to AgNPs in acidic soil (pH < 5.5), while elevated OM content (>1.9 %) attenuated AgNPs toxicity. Compared to soil acidification, reducing soil OM content is more detrimental in exacerbating AgNPs' toxicity and it emerged that clay particles were deemed effective in curbing their toxicity. Meanwhile sand particles played a very different role, and a sandy soil sample at > 40 % of the water holding capacity (WHC), amplified the toxicity of AgNPs. Perturbation mapping of how soil texture alters enzyme activity under AgNPs exposure was generated, where soils with sand (45-65 %), silt (< 22 %), and clay (35-55 %) exhibited even higher probability of positive effects of AgNPs. The average calculation results indicate the sandy clay loam (75.6 %), clay (74.8 %), silt clay (65.8 %), and sandy clay (55.9 %) texture soil demonstrate less AgNPs inhibition effect. The results herein advance the prediction of the effect of AgNPs on soil enzymes globally and determine the soil types that are more sensitive to AgNPs worldwide.

New Insights on the Formation of Nucleation Mode Particles in a Coastal City Based on a Machine Learning Approach

Atmospheric particles have profound implications for the global climate and human health. Among them, ultrafine particles dominate in terms of the number concentration and exhibit enhanced toxic effects as a result of their large total surface area. Therefore, understanding the driving factors behind ultrafine particle behavior is crucial. Machine learning (ML) provides a promising approach for handling complex relationships. In this study, three ML models were constructed on the basis of field observations to simulate the particle number concentration of nucleation mode (PNCN). All three models exhibited robust PNCN reproduction (R2 > 0.80), with the random forest (RF) model excelling on the test data (R2 = 0.89). Multiple methods of feature importance analysis revealed that ultraviolet (UV), H2SO4, low-volatility oxygenated organic molecules (LOOMs), temperature, and O3 were the primary factors influencing PNCN. Bivariate partial dependency plots (PDPs) indicated that during nighttime and overcast conditions, the presence of H2SO4 and LOOMs may play a crucial role in influencing PNCN. Additionally, integrating additional detailed information related to emissions or meteorology would further enhance the model performance. This pilot study shows that ML can be a novel approach for simulating atmospheric pollutants and contributes to a better understanding of the formation and growth mechanisms of nucleation mode particles.