• Bulletin of the Korean Chemical Society
• /
• v.33 no.2
• /
• pp.613-619
• /
• 2012

We applied several machine learning methods for developing QSAR models for prediction of acute toxicity to fathead minnow. The multiple linear regression (MLR) and artificial neural network (ANN) method were applied to predict 96 h $LC_{50}$ (median lethal concentration) of 555 chemical compounds. Molecular descriptors based on 2D chemical structure were calculated by PreADMET program. The recursive partitioning (RP) model was used for grouping of mode of actions as reactive or narcosis, followed by MLR method of chemicals within the same mode of action. The MLR, ANN, and two RP-MLR models possessed correlation coefficients ($R^2$) as 0.553, 0.618, 0.632, and 0.605 on test set, respectively. The consensus model of ANN and two RP-MLR models was used as the best model on training set and showed good predictivity ($R^2$=0.663) on the test set.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.29 no.2
• /
• pp.141-158
• /
• 2019

Objectives: An adverse outcome pathway is a biological pathway that disturbs homeostasis and causes toxicity. It is a conceptual framework for organizing existing biological knowledge and consists of the molecular initiating event, key event, and adverse output. The AOP concept provides intuitive risk identification that can be helpful in evaluating the carcinogenicity of chemicals and in the prevention of cancer through the assessment of chemical carcinogenicity predictions. Methods: We reviewed various papers and books related to the application of AOPs for the prevention of occupational cancer. We mainly used the internet to search for the necessary research data and information, such as via Google scholar(http://scholar.google.com), ScienceDirect(www.sciencedirect.com), Scopus(www.scopus. com), NDSL(http: //www.ndsl.kr/index.do) and PubMed(http://www.ncbi.nlm.nih.gov/pubmed). The key terms searched were "adverse outcome pathway," "toxicology," "risk assessment," "human exposure," "worker," "nanoparticle," "applications," and "occupational safety and health," among others. Results: Since it focused on the current state of AOP for the prediction of toxicity from chemical exposure at work and prospects for industrial health in the context of the AOP concept, respiratory and nanomaterial hazard assessments. AOP provides an intuitive understanding of the toxicity of chemicals as a conceptual means, and it works toward accurately predicting chemical toxicity. The AOP technique has emerged as a future-oriented alternative to the existing paradigm of chemical hazard and risk assessment. AOP can be applied to the assessment of chemical carcinogenicity along with efforts to understand the effects of chronic toxic chemicals in workplaces. Based on these predictive tools, it could be possible to bring about a breakthrough in the prevention of occupational and environmental cancer. Conclusions: The AOP tool has emerged as a future-oriented alternative to the existing paradigm of chemical hazard and risk assessment and has been widely used in the field of chemical risk assessment and the evaluation of carcinogenicity at work. It will be a useful tool for prediction, and it is possible that it can help bring about a breakthrough in the prevention of occupational and environmental cancer.

• Journal of Environmental Health Sciences
• /
• v.45 no.5
• /
• pp.457-464
• /
• 2019

Objectives: Quantitative structure-activity relationship (QSAR) is one of the effective alternatives to animal testing, but its credibility in terms of toxicity prediction has been questionable. Thus, this work aims to evaluate its predictive capacity and find ways of improving its credibility. Methods: Using $TOPKAT^{(R)}$, OECD toolbox, and $Derek^{(R)}$, all of which have been applied world-wide in the research, industrial, and regulatory fields, an analysis of prediction credibility markers including accuracy (A), sensitivity (S), specificity (SP), false negative (FN), and false positive (FP) was conducted. Results: The multi-application of QSARs elevated the precision credibility relative to individual applications of QSARs. Moreover, we found that the type of chemical structure affects the credibility of markers significantly. Conclusions: The credibility of individual QSAR is insufficient for both the prediction of chemical toxicity and regulation of hazardous chemicals. Thus, to increase the credibility, multi-QSAR application, and compensation of the prediction deviation by chemical structure are required.

• Environmental Analysis Health and Toxicology
• /
• v.30 no.sup
• /
• pp.7.1-7.10
• /
• 2015

Objectives For successful adoption of legislation controlling registration and assessment of chemical substances, it is important to obtain sufficient toxicological experimental evidence and other related information. It is also essential to obtain a sufficient number of predicted risk and toxicity results. Particularly, methods used in predicting toxicities of chemical substances during acquisition of required data, ultimately become an economic method for future dealings with new substances. Although the need for such methods is gradually increasing, the-required information about reliability and applicability range has not been systematically provided. Methods There are various representative environmental and human toxicity models based on quantitative structure-activity relationships (QSAR). Here, we secured the 10 representative QSAR-based prediction models and its information that can make predictions about substances that are expected to be regulated. We used models that predict and confirm usability of the information expected to be collected and submitted according to the legislation. After collecting and evaluating each predictive model and relevant data, we prepared methods quantifying the scientific validity and reliability, which are essential conditions for using predictive models. Results We calculated predicted values for the models. Furthermore, we deduced and compared adequacies of the models using the Alternative non-testing method assessed for Registration, Evaluation, Authorization, and Restriction of Chemicals Substances scoring system, and deduced the applicability domains for each model. Additionally, we calculated and compared inclusion rates of substances expected to be regulated, to confirm the applicability. Conclusions We evaluated and compared the data, adequacy, and applicability of our selected QSAR-based toxicity prediction models, and included them in a database. Based on this data, we aimed to construct a system that can be used with predicted toxicity results. Furthermore, by presenting the suitability of individual predicted results, we aimed to provide a foundation that could be used in actual assessments and regulations.

• Journal of Environmental Health Sciences
• /
• v.49 no.2
• /
• pp.89-98
• /
• 2023

Background: Organophosphorus flame retardants (OPFRs) are a group of chemical substances used in building materials and plastic products to suppress or mitigate the combustion of materials. Although OPFRs are generally used in mixed form, information on their mixture toxicity is quite scarce. Objectives: This study aims to elucidate the toxicity and determine the types of interaction (e.g., synergistic, additive, and antagonistic effect) of OPFRs mixtures. Methods: Nine organophosphorus flame retardants, including TEHP (tris(2-ethylhexyl) phosphate) and TDCPP (tris(1,3-dichloro-2-propyl) phosphate), were selected based on indoor dust measurement data in South Korea. Nine OPFRs were exposed to the luminescent bacteria Aliivibrio fischeri for 30 minutes and the human hepatocyte cell line HepG2 for 48 hours. Chemicals with significant toxicity were only used for mixture toxicity tests in HepG2. In addition, the observed EC_x values were compared with the predicted toxicity values in the CA (concentration addition) prediction model, and the MDR (model deviation ratio) was calculated to determine the type of interaction. Results: Only four chemicals showed significant toxicity in the luminescent bacteria assays. However, EC₅₀ values were derived for seven out of nine OPFRs in the HepG2 assays. In the HepG2 assays, the highest to lowest EC₅₀ were in the order of the molecular weight of the target chemicals. In the further mixture tests, most binary mixtures show additive interactions except for the two combinations that have TPhP (triphenyl phosphate), i.e., TPhP and TDCPP, and TPhP and TBOEP (tris(2-butoxyethyl) phosphate). Conclusions: Our data shows OPFR mixtures usually have additivity; however, more research is needed to find out the reason for the synergistic effect of TPhP. Also, the mixture experimental dataset can be used as a training and validation set for developing the mixture toxicity prediction model as a further step.

• Molecular & Cellular Toxicology
• /
• v.1 no.1
• /
• pp.13-16
• /
• 2005

The current vision of toxicogenomics is the development of methods or platforms to predict toxicity of un characterized chemicals by using '-omics' information in pre-clinical stage. Because each chemical has different ADME (absorption, distribution, mechanism, excretion) and experimental animals have lots of variation, precise prediction of chemical's toxicity based on '-omics' information and toxicity data of known chemicals is very difficult problem. So, the importance of bioinformatics is more emphasized on toxicogenomics than other functional genomics studies because these problems can not be solved only with experiments. Thus, toxicoinformatics covers all information-based analytical methods from gene expression (bioinformatics) to chemical structures (cheminformatics) and it also deals with the integration of wide range of experimental data for further extensive analyses. In this review, the overall strategy to toxicoinformatics is discussed.

• Bulletin of the Korean Chemical Society
• /
• v.35 no.12
• /
• pp.3637-3641
• /
• 2014

In this study, we analyzed the toxicity of mixtures of dimethylformamide (DMF) and methyl ethyl ketone (MEK) or DMF and toluene (TOL) and predicted their toxicity using quantitative structure-activity relationships (QSAR). A QSAR model for single substances and mixtures was analyzed using multiple linear regression (MLR) by taking into account the statistical parameters between the observed and predicted $EC_{50}$. After preprocessing, the best subsets of descriptors in the learning methods were determined using a 5-fold cross-validation method. Significant differences in physico-chemical properties such as boiling point (BP), specific gravity (SG), Reid vapor pressure (rVP), flash point (FP), low explosion limit (LEL), and octanol/water partition coefficient (Pow) were observed between the single substances and the mixtures. The $EC_{50}$ of the mixture of DMF and TOL was significantly lower than that of DMF. The mixture toxicity was directly related to the mixing ratio of TOL and MEK (MLR $EC_{50}$ equation = $1.76997-1.12249{\times}TOL+1.21045{\times}MEK$), as well as to SG, VP, and LEL (MLR equation $EC_{50}=15.44388-19.84549{\times}SG+0.05091{\times}VP+1.85846{\times}LEL$). These results show that QSAR-based models can be used to quantitatively predict the toxicity of mixtures used in manufacturing industries.

• Journal of Environmental Health Sciences
• /
• v.39 no.2
• /
• pp.130-137
• /
• 2013

Objectives: The OECD QSAR Application Toolbox was developed by the Organisation for Economic Cooperation and Development (OECD) to facilitate the practical use of QSAR approaches in regulatory contexts as well as to reduce the need for additional animal testing. In this study, human health and the ecotoxicity of chemicals were predicted by applying the OECD QSAR Application Toolbox and the results were compared with experimental data in order to evaluate the applicability of this program. Methods: Read-across, trend analysis, and QSAR of OECD QSAR Application Toolbox were used for the prediction of toxicity. Results: The toxicity prediction was conducted on 6,354 chemicals for which toxicity data have been produced on the six endpoints of skin sensitization, skin irritation, eye irritation, mutagenicity, and acute toxicities of fish and Daphnia. From the total of 6,354, we obtained prediction results for 1,621 chemicals (25.5%). Conclusions: The predicted properties of mutagenicity, skin sensitization, and acute aquatic toxicities were reasonably good when compared with experimental data, but other endpoints were not due to the limitation of applicable chemical groups.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.31 no.3
• /
• pp.185-193
• /
• 2021

Objectives: Inhalation toxicity testing of chemical substances to identify carcinogenicity requires a long time and considerable cost, so the selection of test candidates is a very important aspect. This study was performed to determine optimal procedures for selecting carcinogenic inhalation toxicity test substances as conducted by the Occupational Safety and Health Research Institute (OSHRI). Methods: At the beginning, a database was constructed containing complex information such as usage amount, hazard, carcinogenicity prediction, and testability in order to select chemicals requiring carcinogenicity testing. Selection of test substances was carried out with priority given to usage, carcinogenicity, and testability. Results: Chemicals used in large quantities in industrial fields and strongly suspected of carcinogenicity were winnowed down to 12 substances, and these substances were scheduled for future testing by OSHRI. Conclusions: For the stable and reliable operation of carcinogenicity tests as conducted by OSHRI, this study standardized the procedures for selecting carcinogenicity test substances and suggested the introduction of various carcinogenicity prediction techniques.

• Journal of the Korean Society of Safety
• /
• v.38 no.6
• /
• pp.60-71
• /
• 2023

The purpose of this study is to propose a quantitative toxicity endpoint distance suitable for the initial response of firefighters by comparing and analyzing the commonly applied toxic level of concern (T-LOC), specifically emergency response planning guidelines (ERPG), acute exposure guideline levels (AEGL), and immediately dangerous to life or health (IDLH). This is to protect the fire brigade, which responds to toxic chemical accidents first during the golden time. Using areal locations of hazardous atmospheres, a damage prediction program, the amount of leakage for both acidic and basic substances, along with the endpoint distance, were analyzed for alternative accident and worst-case accident scenarios. The results showed that the toxicity endpoint distance, serving as a compromise between Level-3 and Level-2 of T-LOC, was longer than ERPG-3 and shorter than ERPG-2 with IDLH, while its values were analyzed in the order of ERPG-2, AEGL-2, IDLH, AEGL-3, and ERPG-3. It is suggested that the application of IDLH in an emergency (red card) and ERPG-2 endpoint distance in a non-emergency (non-red card) can be utilized for the initial response of the fire brigade.