1
|
Ji Z, Guo W, Wood EL, Liu J, Sakkiah S, Xu X, Patterson TA, Hong H. Machine Learning Models for Predicting Cytotoxicity of Nanomaterials. Chem Res Toxicol 2022; 35:125-139. [PMID: 35029374 DOI: 10.1021/acs.chemrestox.1c00310] [Citation(s) in RCA: 15] [Impact Index Per Article: 7.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/28/2022]
Abstract
The wide application of nanomaterials in consumer and medical products has raised concerns about their potential adverse effects on human health. Thus, more and more biological assessments regarding the toxicity of nanomaterials have been performed. However, the different ways the evaluations were performed, such as the utilized assays, cell lines, and the differences of the produced nanoparticles, make it difficult for scientists to analyze and effectively compare toxicities of nanomaterials. Fortunately, machine learning has emerged as a powerful tool for the prediction of nanotoxicity based on the available data. Among different types of toxicity assessments, nanomaterial cytotoxicity was the focus here because of the high sensitivity of cytotoxicity assessment to different treatments without the need for complicated and time-consuming procedures. In this review, we summarized recent studies that focused on the development of machine learning models for prediction of cytotoxicity of nanomaterials. The goal was to provide insight into predicting potential nanomaterial toxicity and promoting the development of safe nanomaterials.
Collapse
Affiliation(s)
- Zuowei Ji
- National Center for Toxicological Research, United States Food and Drug Administration, Jefferson, Arkansas 72079, United States
| | - Wenjing Guo
- National Center for Toxicological Research, United States Food and Drug Administration, Jefferson, Arkansas 72079, United States
| | - Erin L Wood
- Office of Pharmaceutical Quality, Center for Drug Evaluation and Research, United States Food and Drug Administration, Silver Spring, Maryland 20993, United States
| | - Jie Liu
- National Center for Toxicological Research, United States Food and Drug Administration, Jefferson, Arkansas 72079, United States
| | - Sugunadevi Sakkiah
- National Center for Toxicological Research, United States Food and Drug Administration, Jefferson, Arkansas 72079, United States
| | - Xiaoming Xu
- Office of Testing and Research, Office of Pharmaceutical Quality, Center for Drug Evaluation and Research, United States Food and Drug Administration, Silver Spring, Maryland 20993, United States
| | - Tucker A Patterson
- National Center for Toxicological Research, United States Food and Drug Administration, Jefferson, Arkansas 72079, United States
| | - Huixiao Hong
- National Center for Toxicological Research, United States Food and Drug Administration, Jefferson, Arkansas 72079, United States
| |
Collapse
|
2
|
Mao J, Akhtar J, Zhang X, Sun L, Guan S, Li X, Chen G, Liu J, Jeon HN, Kim MS, No KT, Wang G. Comprehensive strategies of machine-learning-based quantitative structure-activity relationship models. iScience 2021; 24:103052. [PMID: 34553136 PMCID: PMC8441174 DOI: 10.1016/j.isci.2021.103052] [Citation(s) in RCA: 38] [Impact Index Per Article: 12.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/21/2022] Open
Abstract
Early quantitative structure-activity relationship (QSAR) technologies have unsatisfactory versatility and accuracy in fields such as drug discovery because they are based on traditional machine learning and interpretive expert features. The development of Big Data and deep learning technologies significantly improve the processing of unstructured data and unleash the great potential of QSAR. Here we discuss the integration of wet experiments (which provide experimental data and reliable verification), molecular dynamics simulation (which provides mechanistic interpretation at the atomic/molecular levels), and machine learning (including deep learning) techniques to improve QSAR models. We first review the history of traditional QSAR and point out its problems. We then propose a better QSAR model characterized by a new iterative framework to integrate machine learning with disparate data input. Finally, we discuss the application of QSAR and machine learning to many practical research fields, including drug development and clinical trials.
Collapse
Affiliation(s)
- Jiashun Mao
- The Interdisciplinary Graduate Program in Integrative Biotechnology and Translational Medicine, Yonsei University, Incheon 21983, Republic of Korea
- Department of Biology, School of Life Sciences, Southern University of Science and Technology, 1088 Xueyuan Avenue, Shenzhen, Guangdong 518055, China
- Guangdong Provincial Key Laboratory of Computational Science and Material Design, Shenzhen, Guangdong 518055 China
| | - Javed Akhtar
- Department of Biology, School of Life Sciences, Southern University of Science and Technology, 1088 Xueyuan Avenue, Shenzhen, Guangdong 518055, China
- Guangdong Provincial Key Laboratory of Cell Microenvironment and Disease Research, Shenzhen, Guangdong 518055, China
| | - Xiao Zhang
- Shanghai Rural Commercial Bank Co., Ltd, Shanghai 200002, China
| | - Liang Sun
- Department of Physics, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong, China
| | - Shenghui Guan
- Department of Biology, School of Life Sciences, Southern University of Science and Technology, 1088 Xueyuan Avenue, Shenzhen, Guangdong 518055, China
- Guangdong Provincial Key Laboratory of Computational Science and Material Design, Shenzhen, Guangdong 518055 China
| | - Xinyu Li
- School of Life and Health Sciences and Warshel Institute for Computational Biology, The Chinese University of Hong Kong, Shenzhen 518172, China
| | - Guangming Chen
- Department of Biology, School of Life Sciences, Southern University of Science and Technology, 1088 Xueyuan Avenue, Shenzhen, Guangdong 518055, China
- Guangdong Provincial Key Laboratory of Cell Microenvironment and Disease Research, Shenzhen, Guangdong 518055, China
| | - Jiaxin Liu
- Biotechnology, College of Life Science and Biotechnology, Yonsei University, Seoul 03722, Republic of Korea
| | - Hyeon-Nae Jeon
- Biotechnology, College of Life Science and Biotechnology, Yonsei University, Seoul 03722, Republic of Korea
| | - Min Sung Kim
- Biotechnology, College of Life Science and Biotechnology, Yonsei University, Seoul 03722, Republic of Korea
| | - Kyoung Tai No
- The Interdisciplinary Graduate Program in Integrative Biotechnology and Translational Medicine, Yonsei University, Incheon 21983, Republic of Korea
| | - Guanyu Wang
- Department of Biology, School of Life Sciences, Southern University of Science and Technology, 1088 Xueyuan Avenue, Shenzhen, Guangdong 518055, China
- Guangdong Provincial Key Laboratory of Computational Science and Material Design, Shenzhen, Guangdong 518055 China
- Guangdong Provincial Key Laboratory of Cell Microenvironment and Disease Research, Shenzhen, Guangdong 518055, China
| |
Collapse
|
3
|
Engin AB. Combined Toxicity of Metal Nanoparticles: Comparison of Individual and Mixture Particles Effect. ADVANCES IN EXPERIMENTAL MEDICINE AND BIOLOGY 2021; 1275:165-193. [PMID: 33539016 DOI: 10.1007/978-3-030-49844-3_7] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 01/07/2023]
Abstract
Toxicity of metal nanoparticles (NPs) are closely associated with increasing intracellular reactive oxygen species (ROS) and the levels of pro-inflammatory mediators. However, NP interactions and surface complexation reactions alter the original toxicity of individual NPs. To date, toxicity studies on NPs have mostly been focused on individual NPs instead of the combination of several species. It is expected that the amount of industrial and highway-acquired NPs released into the environment will further increase in the near future. This raises the possibility that various types of NPs could be found in the same medium, thereby, the adverse effects of each NP either could be potentiated, inhibited or remain unaffected by the presence of the other NPs. After uptake of NPs into the human body from various routes, protein kinases pathways mediate their toxicities. In this context, family of mitogen-activated protein kinases (MAPKs) is mostly efficient. Despite each NP activates almost the same metabolic pathways, the toxicity induced by a single type of NP is different than the case of co-exposure to the combined NPs. The scantiness of toxicological data on NPs combinations displays difficulties to determine, if there is any risk associated with exposure to combined nanomaterials. Currently, in addition to mathematical analysis (Response surface methodology; RSM), the quantitative-structure-activity relationship (QSAR) is used to estimate the toxicity of various metal oxide NPs based on their physicochemical properties and levels applied. In this chapter, it is discussed whether the coexistence of multiple metal NPs alter the original toxicity of individual NP. Additionally, in the part of "Toxicity of diesel emission/exhaust particles (DEP)", the known individual toxicity of metal NPs within the DEP is compared with the data regarding toxicity of total DEP mixture.
Collapse
Affiliation(s)
- Ayse Basak Engin
- Department of Toxicology, Faculty of Pharmacy, Gazi University, Ankara, Turkey.
| |
Collapse
|
4
|
Kar S, Pathakoti K, Tchounwou PB, Leszczynska D, Leszczynski J. Evaluating the cytotoxicity of a large pool of metal oxide nanoparticles to Escherichia coli: Mechanistic understanding through In Vitro and In Silico studies. CHEMOSPHERE 2021; 264:128428. [PMID: 33022504 PMCID: PMC7919734 DOI: 10.1016/j.chemosphere.2020.128428] [Citation(s) in RCA: 12] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/01/2020] [Revised: 08/23/2020] [Accepted: 09/21/2020] [Indexed: 05/25/2023]
Abstract
The toxic effect of eight metal oxide nanoparticles (MONPs) on Escherichia coli was experimentally evaluated following standard bioassay protocols. The obtained cytotoxicity ranking of these studied MONPs is Er2O3, Gd2O3, CeO2, Co2O3, Mn2O3, Co3O4, Fe3O4/WO3 (in descending order). The computed EC50 values from experimental data suggested that Er2O3 and Gd2O3 were the most acutely toxic MONPs to E. coli. To identify the mechanism of toxicity of these 8 MONPs along with 17 other MONPs from our previous study, we employed seven classifications and machine learning (ML) algorithms including linear discriminant analysis (LDA), naïve bayes (NB), multinomial logistic regression (MLogitR), sequential minimal optimization (SMO), AdaBoost, J48, and random forest (RF). We also employed 1st and 2nd generation periodic table descriptors developed by us (without any sophisticated computing facilities) along with experimentally analyzed Zeta-potential, to model the cytotoxicity of these MONPs. Based on qualitative validation metrics, the LDA model appeared to be the best among the 7 tested models. The core environment of metal defined by the ratio of the number of core electrons to the number of valence electrons and the electronegativity count of oxygen showed a positive impact on toxicity. The identified properties were important for understanding the mechanisms of nanotoxicity and for predicting the potential environmental risk associated with MONPs exposure. The developed models can be utilized for environmental risk assessment of any untested MONP to E. coli, thereby providing a scientific basis for the design and preparation of safe nanomaterials.
Collapse
Affiliation(s)
- Supratik Kar
- Interdisciplinary Center for Nanotoxicity, Department of Chemistry, Physics and Atmospheric Sciences, Jackson State University, Jackson, MS, 39217, USA
| | - Kavitha Pathakoti
- Interdisciplinary Center for Nanotoxicity, Department of Chemistry, Physics and Atmospheric Sciences, Jackson State University, Jackson, MS, 39217, USA; RCMI Center for Environmental Health, Department of Biology, Jackson State University, Jackson, MS, 39217, USA
| | - Paul B Tchounwou
- Interdisciplinary Center for Nanotoxicity, Department of Chemistry, Physics and Atmospheric Sciences, Jackson State University, Jackson, MS, 39217, USA; RCMI Center for Environmental Health, Department of Biology, Jackson State University, Jackson, MS, 39217, USA
| | - Danuta Leszczynska
- Interdisciplinary Center for Nanotoxicity, Department of Chemistry, Physics and Atmospheric Sciences, Jackson State University, Jackson, MS, 39217, USA; Department of Civil and Environmental Engineering, Jackson State University, Jackson, MS, 39217, USA
| | - Jerzy Leszczynski
- Interdisciplinary Center for Nanotoxicity, Department of Chemistry, Physics and Atmospheric Sciences, Jackson State University, Jackson, MS, 39217, USA.
| |
Collapse
|
5
|
Shin HK, Kim S, Yoon S. Use of size-dependent electron configuration fingerprint to develop general prediction models for nanomaterials. NANOIMPACT 2021; 21:100298. [PMID: 35559785 DOI: 10.1016/j.impact.2021.100298] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Subscribe] [Scholar Register] [Received: 10/17/2020] [Revised: 01/18/2021] [Accepted: 01/18/2021] [Indexed: 06/15/2023]
Abstract
Due to the lack of nano descriptors that can appropriately represent the wide chemical space of engineered nanomaterials (ENMs), applicability domain of nano-quantitative structure-activity relationship models are limited to certain types of ENMs, such as metal oxides, metals, carbon-based nanomaterials, or quantum dots. In this study, a size-dependent electron configuration fingerprint (SDEC FP) was introduced to estimate the quantity of electrons based on the core, doping, and coating materials of ENMs in different sizes. SDEC FP was used in prediction model development and nanostructure similarity analysis on datasets including metal and carbon-based nanomaterials with and without surface modifications. Cytotoxicity and zeta potential prediction models developed with SDEC FP achieved good prediction accuracies on test set. Nanostructure similarity analysis was performed through principal component analysis which showed that structural similarity between ENMs measured by SDEC FP was highly correlated with their properties.
Collapse
Affiliation(s)
- Hyun Kil Shin
- Toxicoinformatics Group, Department of Predictive Toxicology, Korea Institute of Toxicology, Daejeon 34114, Republic of Korea.
| | - Soojin Kim
- Molecular Toxicology Group, Department of Predictive Toxicology, Korea Institute of Toxicology, Daejeon 34114, Republic of Korea
| | - Seokjoo Yoon
- Molecular Toxicology Group, Department of Predictive Toxicology, Korea Institute of Toxicology, Daejeon 34114, Republic of Korea
| |
Collapse
|
6
|
Forest V, Hochepied JF, Pourchez J. Importance of Choosing Relevant Biological End Points To Predict Nanoparticle Toxicity with Computational Approaches for Human Health Risk Assessment. Chem Res Toxicol 2019; 32:1320-1326. [PMID: 31243983 DOI: 10.1021/acs.chemrestox.9b00022] [Citation(s) in RCA: 21] [Impact Index Per Article: 4.2] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 01/08/2023]
Abstract
Because it is impossible to assess in vitro or in vivo the toxicity of all nanoparticles available on the market on a case-by-case basis, computational approaches have been proposed as useful alternatives to predict in silico the hazard potential of engineered nanoparticles. Despite promising results, a major issue associated with these mathematical models lies in the a priori choice of the physicochemical descriptors and the biological end points. We performed a thorough bibliographic survey on the biological end points used for nanotoxicology purposes and compared them between experimental and computational approaches. They were found to be disparate: while conventional in vitro nanotoxicology assays usually investigate a large array of biological effects using eukaryotic cells (cytotoxicity, pro-inflammatory response, oxidative stress, genotoxicity), computational studies mostly focus on cell viability and also include studies on prokaryotic cells. We may thus wonder the relevance of building complex mathematical models able to predict accurately a biological end point if this latter is not the most relevant to support human health risk assessment. The choice of biological end points clearly deserves to be more carefully discussed. This could bridge the gap between experimental and computational nanotoxicology studies and allow in silico predictive models to reach their full potential.
Collapse
Affiliation(s)
- Valérie Forest
- Mines Saint-Etienne, Univ Lyon, Univ Jean Monnet , INSERM, U 1059 Sainbiose, Centre CIS , F-42023 Saint-Etienne , France
| | - Jean-François Hochepied
- MINES ParisTech , PSL Research University , MAT - Centre des matériaux, CNRS UMR 7633 , BP 87 91003 Evry , France.,UCP, ENSTA ParisTech , Université Paris-Saclay , 828 bd des Maréchaux , 91762 Palaiseau cedex , France
| | - Jérémie Pourchez
- Mines Saint-Etienne, Univ Lyon, Univ Jean Monnet , INSERM, U 1059 Sainbiose, Centre CIS , F-42023 Saint-Etienne , France
| |
Collapse
|