-
工程纳米颗粒(engineered nanoparticles, ENPs)是指至少在一维尺度上介于1—100 nm的颗粒,凭借其优异的理化性质得到各个领域的“青睐”,广泛用于个人护理产品、油漆、催化剂、半导体、医学成像以及太阳能电池等[1 − 5]. 随着纳米技术的不断发展,工程纳米材料的产量迅猛增加,从2010年的21千吨增长到2020年的58千吨[6]. 然而,ENPs的生产和使用,不可避免地导致其暴露在环境中,并与生态系统中不同营养级生物相互作用,引起了人们对ENPs生物暴露和潜在风险的担忧[7 − 8]. 水环境是ENPs的一个重要归宿,ENPs会通过地下水修复、污水排放等各个渠道进入其中[9 − 12]. 进入水环境中的ENPs会被生物摄入并积累,进而产生毒性效应,如DNA损伤、产生氧化应激、产生组织病理学变化以及生长受到抑制等[13 − 21]. 有研究表明,只有被摄入并积累的ENPs才是对水生生物产生毒性效应的关键[22]. 因此,有必要研究水生生物对ENPs的积累,以提供毒理学研究所必需的信息.
负荷量(body burden, BB)代表给定时间生物体内的ENPs浓度,是定量ENPs生物积累的基础参数[23 − 25]. 现有的研究多通过化学消解、放射性元素标记或者荧光标记等实验方法测定ENPs的负荷量,过程复杂且耗时耗力[19, 24, 26]. ENPs的种类繁多,很难通过实验来满足毒理学研究对ENPs负荷量数据的需要. 因此,亟需发展可代替传统实验方法的ENPs负荷量预测模型. 目前,关于ENPs生物积累的预测模型的研究十分有限[23, 27]. Dong等[23]应用机器学习(ML)算法对不同ENPs的负荷量进行预测,以此来满足毒代动力学模型的数据需求,实现对ENPs生物积累潜力的评估. 然而,现有模型对BB的预测是在个体水平上的[23]. ENPs的生物积累具有组织器官依赖性,了解ENPs在生物体内的分布,以及在关键靶器官中的生物积累尤为重要[27 − 28]. 此外,现有的许多ML算法是一些“黑箱模型”,如随机森林,极端梯度提升等[29]. 因此,除了要关注模型的预测性能外,还需要考虑模型的可解释性. SHAP (SHapley Additive exPlanations)方法整合了几种具有坚实理论背景的方法(如局部可解释的模型-诊断性解释、博弈论),并提供了强大的估计算法和软件,它与机器学习的结合成为识别特征重要性以及探索数据关系的有力工具[29 − 31].
本研究构建了涵盖17种ENPs和23种水生生物的BB数据集(n =
1303 ). 基于该数据集,将ENPs的本征特性、暴露参数、生物物种以及组织器官类型等43个参数作为描述符,综合考虑前人研究以及不同机器学习算法的适应性,采用多元线性回归、支持向量机、随机森林、梯度提升决策树和极端梯度提升算法建立了ENPs的BB定量预测模型. 本研究通过增加数据集中生物种类,扩大了模型应用域,实现了特定暴露时间、暴露浓度下生物体内ENPs负荷量的预测,并将负荷量的预测精准到组织器官水平. 此外,为了弥补机器学习算法在可解释性方面的不足,本研究结合SHAP方法解析不同特征对ENPs负荷量的影响,拓展了对ENPs生物积累的认识与理解.
工程纳米颗粒生物负荷量的机器学习预测模型
Machine learning prediction model on body burden of engineered nanoparticles in organisms
-
摘要: 工程纳米颗粒(ENPs)在生物体内的动态浓度即负荷量(BB),是评估ENPs生物积累的基础参数. 大多数研究都通过实验手段来测定BB,其过程繁琐复杂. 构建BB预测模型成为代替动物实验的有效途径. 现有预测模型涵盖的ENPs以及生物种类较少,而且是基于生物整体的BB数据构建的,很难用于预测关键组织器官的负荷情况. 了解ENPs在生物体内的分布,以及在关键靶器官中的积累尤为重要. 本研究通过文献挖掘,构建了涵盖17种ENPs以及23种水生生物的BB数据集(n =
1303 ),采用5种机器学习算法构建模型,并结合SHAP方法解析了不同特征对lgBB的影响. 结果表明,非线性模型要优于线性模型,极端梯度提升(XGBoost)模型的效果最佳(R2adj-train = 0.971,Q2test = 0.909,Q210-CV = 0.887). 通过模型特征重要性分析,发现ENPs的本征特性(密度、粒径、电负性、分子量)、暴露参数(暴露浓度、暴露时间)、生物种类以及组织器官类型是影响ENPs负荷量的关键因素. 其中,密度是影响ENPs负荷量的首要本征特征,负荷量会随着材料密度的增加而减少. 本研究通过增加数据集中生物种类,扩大了模型应用域,实现特定暴露时间和暴露浓度下ENPs负荷量的预测,并将预测结果精准到组织器官水平(脑、脾脏、肌肉等),为ENPs生物积累的研究提供了有效方法.Abstract: Dynamic concentration of engineered nanoparticles (ENPs) in an organism, commonly known as body burden (BB), is a fundamental parameter for assessing the bioaccumulation of ENPs. Prediction models on BB are effective alternatives to complex and time-consuming experimental methods. However, existing prediction models involved relatively few types of ENPs and biological species, and were based on BB data of whole organisms, making it difficult to predict BB in specific tissues or organs. Furthermore, the tissue-organ dependence of ENPs bioaccumulation highlights the importance of understanding their distribution in organisms and their bioaccumulation in key target organs. In this study, a BB dataset (n =1303 ) covering 17 types of ENPs, 23 aquatic organisms, and 13 different tissue organs was compiled through literature mining. Five machine learning algorithms were used to build models, and SHapley Additive exPlanations (SHAP) analysis was conducted to examine how different features of ENPs affect BB. The modeling results demonstrated that nonlinear models outperformed linear models, and the extreme gradient boosting (XGBoost) model performed best (R2adj-train = 0.971, Q2test = 0.909, Q210-CV = 0.887). SHAP analysis showed that the intrinsic properties of ENPs (density, particle size, electronegativity and molecular weight), exposure parameters (exposure concentrations and time), biological species, and organ or tissues types were key factors affecting BB. Among them, density was identified as the primary basic feature affecting BB, i.e., the BB decreased with increasing material density. By increasing the number of biological species and types of ENPs in the dataset, this study expands the application domain of the model, and achieves the prediction of the BB of ENPs under specific exposure time and exposure concentrations. The model developed in this study could predict BB accurately at the tissue/organ level (brain, spleen, muscle, etc.), which provided an effective method for the assessment of ENPs bioaccumulation.-
Key words:
- engineering nanoparticles /
- bioaccumulation /
- body burden /
- tissue and organ /
- machine learning.
-
表 1 物种名称
Table 1. Name of the species
拉丁学名
Latin scientific name中文名称
Chinese Name拉丁学名
Latin scientific name中文名称
Chinese NameDaphnia magna 大型溞 Capitella teleta 海蠕虫 Artemia salina 卤虫 Capoeta fusca 棕二鬚鲃 Danio rerio 斑马鱼 Carassius auratus 金鱼 Cyprinus carpio 鲤鱼 Ceriodaphnia dubia 网纹溞 Chlorella sp. 小球藻 Oncorhynchus mykiss 虹鳟鱼 Gammarus fossarum 片脚类动物 Leptocheirus plumulosus 片脚类动物 Hyalella azteca 端足虫 Lymnaea stagnalis 静水椎实螺 Mytilus galloprovincialis 紫贻贝 Nereis diversicolo 沙蚕 Oreochromis niloticus 罗非鱼 Procambarus clarkii 克氏原螯虾 Prochilodus lineatus 条纹鲮脂鲤 Planorbarius corneus 角类扁卷螺 Chlamydomonas reinhardtii 莱茵衣藻 Chlorella pyrenoidosa 蛋白核小球藻 Desmodesmus subspicatus 近具棘链带藻 表 2 模型相关统计参数汇总
Table 2. Summary of statistical parameters of models
模型
ModelR2adj-train Q2test Q210-CV RMSEtrain RMSEtest MLR 0.552 0.478 0.523 0.991 1.170 SVM 0.917 0.812 0.834 0.337 0.702 RF 0.917 0.899 0.879 0.249 0.516 GBDT 0.917 0.829 0.796 0.422 0.670 XGBoost 0.971 0.909 0.887 0.232 0.489 表 3 本研究与其他模型的比较
Table 3. Comparison of the current model with previous models
模型
Model特征
数目
Number of
features数据量
Number of
datasetsENPs种类
Types of
ENPs物种数量
Number of
species生物结构层次
Biological
structure
hierarch最优算法
Optimal
algorithmR2train RMSEtrain Q2test RMSEtest Dong等[23] 24 577 11 9 个体 XGBoost 0.98 0.19 — — 本研究 24 1303 17 23 个体/组织器官 XGBoost 0.97 0.23 0.91 0.45 注:Dong等的研究中并未提供关于测试集的预测效果,故无法进行比较.
Note: Prediction ability on test set was not provided in the study of Dong et al. and therefore could not be compared. -
[1] LEAD J R, BATLEY G E, ALVAREZ P J J, et al. Nanomaterials in the environment: Behavior, fate, bioavailability, and effects—An updated review[J]. Environmental Toxicology and Chemistry, 2018, 37(8): 2029-2063. doi: 10.1002/etc.4147 [2] SRIVASTAVA V, GUSAIN D, SHARMA Y C. Critical review on the toxicity of some widely used engineered nanoparticles[J]. Industrial & Engineering Chemistry Research, 2015, 54(24): 6209-6233. [3] CHRISTIAN P, VON DER KAMMER F, BAALOUSHA M, et al. Nanoparticles: Structure, properties, preparation and behaviour in environmental media[J]. Ecotoxicology, 2008, 17(5): 326-343. doi: 10.1007/s10646-008-0213-1 [4] JOO S H, ZHAO D Y. Environmental dynamics of metal oxide nanoparticles in heterogeneous systems: A review[J]. Journal of Hazardous Materials, 2017, 322: 29-47. doi: 10.1016/j.jhazmat.2016.02.068 [5] BATHI J R, MOAZENI F, UPADHYAYULA V K K, et al. Behavior of engineered nanoparticles in aquatic environmental samples: Current status and challenges[J]. Science of the Total Environment, 2021, 793: 148560. doi: 10.1016/j.scitotenv.2021.148560 [6] WU X, WANG W, ZHU L Z. Enhanced organic contaminants accumulation in crops: Mechanisms, interactions with engineered nanomaterials in soil[J]. Environmental Pollution, 2018, 240: 51-59. doi: 10.1016/j.envpol.2018.04.072 [7] ADAM V, WU Q, NOWACK B. Integrated dynamic probabilistic material flow analysis of engineered materials in all European countries[J]. NanoImpact, 2021, 22: 100312. doi: 10.1016/j.impact.2021.100312 [8] VALSAMI-JONES E, LYNCH I. NANOSAFETY. How safe are nanomaterials?[J]. Science, 2015, 350(6259): 388-389. doi: 10.1126/science.aad0768 [9] YANG Y, ZHANG C Q, HU Z Q. Impact of metallic and metal oxide nanoparticles on wastewater treatment and anaerobic digestion[J]. Environ Sci:Processes Impacts, 2013, 15(1): 39-48. doi: 10.1039/C2EM30655G [10] ZHANG F, WANG Z, VIJVER M G, et al. Prediction of the joint toxicity of multiple engineered nanoparticles: The integration of classic mixture models and In silico methods[J]. Chemical Research in Toxicology, 2021, 34(2): 176-178. doi: 10.1021/acs.chemrestox.0c00300 [11] BUNDSCHUH M, FILSER J, LÜDERWALD S, et al. Nanoparticles in the environment: Where do we come from, where do we go to?[J]. Environmental Sciences Europe, 2018, 30(1): 1-17. doi: 10.1186/s12302-017-0129-6 [12] TURAN N B, ERKAN H S, ENGIN G O, et al. Nanoparticles in the aquatic environment: Usage, properties, transformation and toxicity—a review[J]. Process Safety and Environmental Protection, 2019, 130: 238-249. doi: 10.1016/j.psep.2019.08.014 [13] LI F M, LIANG Z, ZHENG X, et al. Toxicity of nano-TiO2 on algae and the site of reactive oxygen species production[J]. Aquatic Toxicology, 2015, 158: 1-13. doi: 10.1016/j.aquatox.2014.10.014 [14] MILLER R J, LENIHAN H S, MULLER E B, et al. Impacts of metal oxide nanoparticles on marine phytoplankton[J]. Environmental Science & Technology, 2010, 44(19): 7329-7334. [15] MATOUKE M M, ELEWA D T, ABDULLAHI K. Binary effect of titanium dioxide nanoparticles (nTiO2) and phosphorus on microalgae (Chlorella ‘Ellipsoides Gerneck, 1907)[J]. Aquatic Toxicology, 2018, 198: 40-48. doi: 10.1016/j.aquatox.2018.02.009 [16] MAHANA A, GULIY O I, MEHTA S K. Accumulation and cellular toxicity of engineered metallic nanoparticle in freshwater microalgae: Current status and future challenges[J]. Ecotoxicology and Environmental Safety, 2021, 208: 111662. doi: 10.1016/j.ecoenv.2020.111662 [17] FAN W H, SHI Z W, YANG X P, et al. Bioaccumulation and biomarker responses of cubic and octahedral Cu2O micro/nanocrystals in Daphnia magna[J]. Water Research, 2012, 46(18): 5981-5988. doi: 10.1016/j.watres.2012.08.019 [18] CHEN X J, ZHU Y, YANG K, et al. Nanoparticle TiO2 size and rutile content impact bioconcentration and biomagnification from algae to daphnia[J]. Environmental Pollution, 2019, 247: 421-430. doi: 10.1016/j.envpol.2019.01.022 [19] XIAO B W, WANG X L, YANG J, et al. Bioaccumulation kinetics and tissue distribution of silver nanoparticles in zebrafish: The mechanisms and influence of natural organic matter[J]. Ecotoxicology and Environmental Safety, 2020, 194: 110454. doi: 10.1016/j.ecoenv.2020.110454 [20] LV X H, YANG Y, TAO Y, et al. A mechanism study on toxicity of graphene oxide to Daphnia magna: Direct link between bioaccumulation and oxidative stress[J]. Environmental Pollution, 2018, 234: 953-959. doi: 10.1016/j.envpol.2017.12.034 [21] CHEN J Y, DONG X, XIN Y Y, et al. Effects of titanium dioxide nano-particles on growth and some histological parameters of zebrafish (Danio rerio) after a long-term exposure[J]. Aquatic Toxicology, 2011, 101(3/4): 493-499. [22] HOU W C, WESTERHOFF P, POSNER J D. Biological accumulation of engineered nanomaterials: A review of current knowledge[J]. Environmental Science:Processes & Impacts, 2013, 15(1): 103-122. [23] DONG S P, WU Z H, WANG M J, et al. Assessing comparable bioconcentration potentials for nanoparticles in aquatic organisms via combined utilization of machine learning and toxicokinetic models[J]. SmartMat, 2023,4(3):e1155. [24] MUNA M, HEINLAAN M, BLINOVA I, et al. Evaluation of the effect of test medium on total Cu body burden of nano CuO-exposed Daphnia magna: A TXRF spectroscopy study[J]. Environmental Pollution, 2017, 231: 1488-1496. doi: 10.1016/j.envpol.2017.07.083 [25] 胡奕, 王艳龙, 林道辉. 纳米颗粒对大型蚤毒性效应的研究进展[J]. 科学通报, 2017, 62(24): 2734-2748. doi: 10.1360/N972017-00476 HU Y, WANG Y L, LIN D H. A review of the toxicity of nanoparticles to Daphnia magna[J]. Chinese Science Bulletin, 2017, 62(24): 2734-2748 (in Chinese). doi: 10.1360/N972017-00476
[26] SKJOLDING L M, KERN K, HJORTH R, et al. Uptake and depuration of gold nanoparticles in Daphnia magna[J]. Ecotoxicology, 2014, 23(7): 1172-1183. doi: 10.1007/s10646-014-1259-x [27] ZHENG Y F, NOWACK B. Meta-analysis of bioaccumulation data for nondissolvable engineered nanomaterials in freshwater aquatic organisms[J]. Environmental Toxicology and Chemistry, 2022, 41(5): 1202-1214. doi: 10.1002/etc.5312 [28] VIJVER M G, ZHAI Y J, WANG Z, et al. Emerging investigator series: The dynamics of particle size distributions need to be accounted for in bioavailability modelling of nanoparticles[J]. Environmental Science:Nano, 2018, 5(11): 2473-2481. doi: 10.1039/C8EN00572A [29] LI Z Q. Extracting spatial effects from machine learning model using local interpretation method: An example of SHAP and XGBoost[J]. Computers, Environment and Urban Systems, 2022, 96: 101845. doi: 10.1016/j.compenvurbsys.2022.101845 [30] PARIMBELLI E, BUONOCORE T M, NICORA G, et al. Why did AI get this one wrong?—Tree-based explanations of machine learning model predictions[J]. Artificial Intelligence in Medicine, 2023, 135: 102471. doi: 10.1016/j.artmed.2022.102471 [31] DOLK D, KRIDEL D, DINEEN J, et al. Model interpretation and explainability towards creating transparency in prediction models[C]//Proceedings of the Annual Hawaii International Conference on System Sciences, 2020. [32] ZHAO C M, WANG W X. Biokinetic uptake and efflux of silver nanoparticles in Daphnia magna[J]. Environmental Science & Technology, 2010, 44(19): 7699-7704. [33] ZHAO C M, WANG W X. Comparison of acute and chronic toxicity of silver nanoparticles and silver nitrate to Daphnia magna[J]. Environmental Toxicology and Chemistry, 2011, 30(4): 885-892. doi: 10.1002/etc.451 [34] KUEHR S, KAEGI R, MALETZKI D, et al. Testing the bioaccumulation potential of manufactured nanomaterials in the freshwater amphipod Hyalella azteca[J]. Chemosphere, 2021, 263: 127961. doi: 10.1016/j.chemosphere.2020.127961 [35] SIKDER M, EUDY E, CAI B, et al. Particle size determines the accumulation of platinum nanoparticles in the estuarine amphipod, Leptocheirus plumulosus[J]. Environmental Science:Nano, 2022, 9(2): 499-510. doi: 10.1039/D1EN00713K [36] ZHANG Y, ZHU L, ZHOU Y, et al. Accumulation and elimination of iron oxide nanomaterials in zebrafish (Danio rerio) upon chronic aqueous exposure[J]. Journal of Environmental Sciences, 2015, 30: 223-230. doi: 10.1016/j.jes.2014.08.024 [37] HEINLAAN M, MUNA M, JUGANSON K, et al. Exposure to sublethal concentrations of Co3O4 and Mn2O3 nanoparticles induced elevated metal body burden in Daphnia magna[J]. Aquatic Toxicology, 2017, 189: 123-133. doi: 10.1016/j.aquatox.2017.06.002 [38] GÜRKAN M, GÜRKAN S E, YıLMAZ S, et al. Comparative toxicity of alpha and gamma iron oxide nanoparticles in rainbow trout: Histopathology, hematology, accumulation, and oxidative stress[J]. Water, Air, & Soil Pollution, 2021, 232(2): 1-14. [39] CONG Y, BANTA G T, SELCK H, et al. Toxicity and bioaccumulation of sediment-associated silver nanoparticles in the estuarine polychaete, Nereis (Hediste) diversicolor[J]. Aquatic Toxicology, 2014, 156: 106-115. doi: 10.1016/j.aquatox.2014.08.001 [40] HULL M S, VIKESLAND P J, SCHULTZ I R. Uptake and retention of metallic nanoparticles in the Mediterranean mussel (Mytilus galloprovincialis)[J]. Aquatic Toxicology, 2013, 140/141: 89-97. doi: 10.1016/j.aquatox.2013.05.005 [41] RIVERO ARZE A, MOUNEYRAC C, CHATEL A, et al. Comparison of uptake and elimination kinetics of metallic oxide nanomaterials on the freshwater microcrustacean Daphnia magna[J]. Nanotoxicology, 2021, 15(9): 1168-1179. doi: 10.1080/17435390.2021.1994668 [42] SAYADI M H, PAVLAKI M D, MARTINS R, et al. Bioaccumulation and toxicokinetics of zinc oxide nanoparticles (ZnO NPs) co-exposed with graphene nanosheets (GNs) in the blackfish (Capoeta fusca)[J]. Chemosphere, 2021, 269: 128689. doi: 10.1016/j.chemosphere.2020.128689 [43] JOO H S, KALBASSI M R, YU I J, et al. Bioaccumulation of silver nanoparticles in rainbow trout (Oncorhynchus mykiss): Influence of concentration and salinity[J]. Aquatic Toxicology, 2013, 140/141: 398-406. doi: 10.1016/j.aquatox.2013.07.003 [44] BENAVIDES M, FERNÁNDEZ-LODEIRO J, COELHO P, et al. Single and combined effects of aluminum (Al2O3) and zinc (ZnO) oxide nanoparticles in a freshwater fish, Carassius auratus[J]. Environmental Science and Pollution Research, 2016, 23(24): 24578-24591. doi: 10.1007/s11356-016-7915-3 [45] ZHAO J, WANG Z Y, LIU X Y, et al. Distribution of CuO nanoparticles in juvenile carp (Cyprinus carpio) and their potential toxicity[J]. Journal of Hazardous Materials, 2011, 197: 304-310. doi: 10.1016/j.jhazmat.2011.09.094 [46] CONG Y, BANTA G T, SELCK H, et al. Toxic effects and bioaccumulation of nano-, micron- and ionic-Ag in the polychaete, Nereis diversicolor[J]. Aquatic Toxicology, 2011, 105(3/4): 403-411. [47] ATES M, ARSLAN Z, DEMIR V, et al. Accumulation and toxicity of CuO and ZnO nanoparticles through waterborne and dietary exposure of goldfish (Carassius auratus)[J]. Environmental Toxicology, 2015, 30(1): 119-128. doi: 10.1002/tox.22002 [48] BEN BRAHIM A, LIMAM M. Ensemble feature selection for high dimensional data: A new method and a comparative study[J]. Advances in Data Analysis and Classification, 2018, 12(4): 937-952. doi: 10.1007/s11634-017-0285-y [49] JIANG C S, ZHAO P P, LI W H, et al. In silico prediction of chemical neurotoxicity using machine learning[J]. Toxicology Research, 2020, 9(3): 164-172. doi: 10.1093/toxres/tfaa016 [50] MURATOV E N, BAJORATH J, SHERIDAN R P, et al. Correction: QSAR without borders[J]. Chemical Society Reviews, 2020, 49(11): 3716. doi: 10.1039/D0CS90041A [51] 丁蕊, 陈景文, 于洋, 等. 基于集成学习算法构建有机化学品鱼体生物富集因子的QSAR预测模型[J]. 环境化学, 2021, 40(5): 1295-1304. doi: 10.7524/j.issn.0254-6108.2021011304 DING R, CHEN J W, YU Y, et al. Using ensemble learning algorithms to develop QSAR models on bioconcentration factors of organic chemicals in multispecies fish[J]. Environmental Chemistry, 2021, 40(5): 1295-1304 (in Chinese). doi: 10.7524/j.issn.0254-6108.2021011304
[52] OUASSIL N, PINALS R L, del BONIS-O’DONNELL J T, et al. Supervised learning model predicts protein adsorption to carbon nanotubes[J]. Science Advances, 2022, 8(1): eabm0898. doi: 10.1126/sciadv.abm0898 [53] JIANG D J, WU Z X, HSIEH C Y, et al. Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models[J]. Journal of Cheminformatics, 2021, 13(1): 1-23. doi: 10.1186/s13321-020-00477-w [54] SVETNIK V, WANG T, TONG C, et al. Boosting: an ensemble learning tool for compound classification and QSAR modeling[J]. Journal of Chemical Information and Modeling, 2005, 45(3): 786-799. doi: 10.1021/ci0500379 [55] SHERIDAN R P, WANG W M, LIAW A, et al. Extreme gradient boosting as a method for quantitative structure–activity relationships[J]. Journal of Chemical Information and Modeling, 2016, 56(12): 2353-2360. doi: 10.1021/acs.jcim.6b00591 [56] WU Z X, ZHU M F, KANG Y, et al. Do we need different machine learning algorithms for QSAR modeling?A comprehensive assessment of 16 machine learning algorithms on 14 QSAR data sets[J]. Briefings in Bioinformatics, 2021, 22(4): bbaa321. doi: 10.1093/bib/bbaa321 [57] HUANG Y, LI X H, XU S J, et al. Quantitative structure–activity relationship models for predicting inflammatory potential of metal oxide nanoparticles[J]. Environmental Health Perspectives, 2020, 128(6): 67010. doi: 10.1289/EHP6508 [58] JAWORSKA J, NIKOLOVA-JELIAZKOVA N, ALDENBERG T. QSAR applicability domain estimation by projection of the training set in descriptor space: A review[J]. Alternatives to Laboratory Animals, 2005, 33(5): 445-459. doi: 10.1177/026119290503300508 [59] GAO Y F, XIE Z C, FENG J F, et al. Different factors determined the toxicokinetics of organic chemicals and nanomaterials exposure to zebrafish (Danio Rerio)[J]. Ecotoxicology and Environmental Safety, 2019, 186: 109810. doi: 10.1016/j.ecoenv.2019.109810 [60] ZHU X S, TIAN S Y, CAI Z H. Toxicity assessment of iron oxide nanoparticles in zebrafish (Danio rerio) early life stages[J]. PLoS One, 2012, 7(9): e46286. doi: 10.1371/journal.pone.0046286 [61] 李宁静, 许喆, 姚烘烨, 等. 粒径对纳米氧化锌的生物累积和氧化应激的影响[J]. 科学通报, 2021, 66(24): 3219-3226. doi: 10.1360/TB-2021-0186 LI N J, XU Z, YAO H Y, et al. Impact of particle size of zinc oxide nanoparticles on its bioaccumulation and oxidative stress responses[J]. Chinese Science Bulletin, 2021, 66(24): 3219-3226 (in Chinese). doi: 10.1360/TB-2021-0186
[62] KLEANDROVA V V, LUAN F, GONZÁLEZ-DÍAZ H, et al. Computational ecotoxicology: Simultaneous prediction of ecotoxic effects of nanoparticles under different experimental conditions[J]. Environment International, 2014, 73: 288-294. doi: 10.1016/j.envint.2014.08.009 [63] ROBINSON R L M, SARIMVEIS H, DOGANIS P, et al. Identifying diverse metal oxide nanomaterials with lethal effects on embryonic zebrafish using machine learning[J]. Beilstein Journal of Nanotechnology, 2021, 12: 1297-1325. doi: 10.3762/bjnano.12.97 [64] ZHU X S, WANG J X, ZHANG X Z, et al. Trophic transfer of TiO2 nanoparticles from daphnia to zebrafish in a simplified freshwater food chain[J]. Chemosphere, 2010, 79(9): 928-933. doi: 10.1016/j.chemosphere.2010.03.022 [65] YUN X Y, LEWIS A J, STEVENS-KING G, et al. Bioaccumulation of per- and polyfluoroalkyl substances by freshwater benthic macroinvertebrates: Impact of species and sediment organic carbon content[J]. Science of the Total Environment, 2023, 866: 161208. doi: 10.1016/j.scitotenv.2022.161208 [66] ROSENKRANZ P, CHAUDHRY Q, STONE V, et al. A comparison of nanoparticle and fine particle uptake by Daphnia magna[J]. Environmental Toxicology and Chemistry, 2009, 28(10): 2142-2149. doi: 10.1897/08-559.1 [67] ABDOLAHPUR MONIKH F, CHUPANI L, ARENAS-LAGO D, et al. Particle number-based trophic transfer of gold nanomaterials in an aquatic food chain[J]. Nature Communications, 2021, 12: 899. doi: 10.1038/s41467-021-21164-w [68] ATES M, DEMIR V, ARSLAN Z, et al. Chronic exposure of tilapia (Oreochromis niloticus) to iron oxide nanoparticles: Effects of particle morphology on accumulation, elimination, hematology and immune responses[J]. Aquatic Toxicology, 2016, 177: 22-32. doi: 10.1016/j.aquatox.2016.05.005