Prediction of Acute Aquatic Toxicity toward <i>Daphnia Magna</i> by using the GA-<i>k</i>NN Method

Cassotti, Matteo; Ballabio, Davide; Consonni, Viviana; Mauri, Andrea; Tetko, Igor V.; Todeschini, Roberto

doi:10.1177/026119291404200106

Cited by 65 publications

(40 citation statements)

References 55 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The MICHEM model appeared to have comparable performance with other QSAR models based on large heterogeneous datasets [2]: the ChemProp model, as well as models proposed by Kar and Roy [61] and Kaiser and Niculescu [62]. The MICHEM model benefited from a simple algorithm, self-determination of its own AD and the possibility of carrying out a chemical analysis for each prediction based on the nearest neighbours.…”

Section: Comparison With Published Modelsmentioning

confidence: 87%

“…Our previous aquatic toxicity model [2], hereinafter referred to as the MICHEM model, was calibrated on a training set of 436 organic molecules and tested on an external set of 110 compounds, randomly selected. Toxicity data were retrieved from three databases (ECOTOX [7], ECETOC [8] and OASIS [9]) and available scientific publications [10][11][12][13][14][15][16][17][18][19][20][21][22][23][24][25][26] The MICHEM model implements a modified kNN approach that uses a threshold on the average Mahalanobis distance from the first three nearest neighbours.…”

Section: Materials and Methods 21 Aquatic Toxicity Modelsmentioning

confidence: 99%

“…This is particularly important for the MICHEM model to verify the assumptions made in the modelling stage regarding the relationship between residuals and average distance from the three nearest neighbours [2]. Figure 4 shows the average Mahalanobis distance of the molecules in the 'External to MICHEM' subset from their three nearest neighbours in the training set versus the standardized residuals, and allows checking of whether the introduced threshold is an effective tool to identify dissimilar molecules whose predictions are likely to be unreliable.…”

Section: Investigation Of the Residuals Of The Michem Modelmentioning

confidence: 99%

“…Among ecotoxicological properties, short-term toxicity to Daphnia is required for all substances subject to REACH. Since REACH promotes the adoption of alternative testing approaches such as in vitro and in silico methods, several (quantitative) structure-activity relationship [(Q)SAR] models were calibrated to address the problem of predicting short-term toxicity to Daphnia magna on both small homogeneous and large heterogeneous datasets [2]. Some models were explicitly developed to comply with the five Organisation for Economic Co-operation and Development (OECD) principles for the validation of QSAR models [3].…”

Section: Introductionmentioning

confidence: 99%

“…Recently, we proposed a QSAR model for the prediction of short-term toxicity (LC 50 48 hours) to D. magna [2]. The entire development process was carried out with the aim to comply with the five OECD principles.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Validation and extension of a similarity-based approach for prediction of acute aquatic toxicity towards Daphnia magna

Cassotti

Consonni

Mauri

et al. 2014

SAR and QSAR in Environmental Research

View full text Add to dashboard Cite

Quantitative structure-activity relationship (QSAR) models for predicting acute toxicity to Daphnia magna are often associated with poor performances, urging the need for improvement to meet REACH requirements. The aim of this study was to evaluate the accuracy, stability and reliability of a previously published QSAR model by means of further external validation and to optimize its performance by means of extension to new data as well as a consensus approach. The previously published model was validated with a large set of new molecules and then compared with ChemProp model, from which most of the validation data were taken. Results showed better performance of the proposed model in terms of accuracy and percentage of molecules outside the applicability domain. The model was re-calibrated on all the available data to confirm the efficacy of the similarity-based approach. The extended dataset was also used to develop a novel model based on the same similarity approach but using binary fingerprints to describe the chemical structures. The fingerprint-based model gave lower regression statistics, but also less unpredicted compounds. Eventually, consensus modelling was successfully used to enhance the accuracy of the predictions and to halve the percentage of molecules outside the applicability domain.

show abstract

Section: Comparison With Published Modelsmentioning

confidence: 87%

Section: Materials and Methods 21 Aquatic Toxicity Modelsmentioning

confidence: 99%

Section: Investigation Of the Residuals Of The Michem Modelmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Validation and extension of a similarity-based approach for prediction of acute aquatic toxicity towards Daphnia magna

Cassotti

Consonni

Mauri

et al. 2014

SAR and QSAR in Environmental Research

View full text Add to dashboard Cite

show abstract

Chemometric Modeling of Daphnia Toxicity

Halder

Cordeiro

2021

Chemometrics and Cheminformatics in Aquatic Toxicology

View full text Add to dashboard Cite

IQR CUSUM charts: An efficient approach for monitoring variations in aquatic toxicity

Hussain

Sun

Mahmood

et al. 2021

Journal of Chemometrics

View full text Add to dashboard Cite

Toxicant testing is a major component of the environmental solution testing program. This test system aims to evaluate the sensitivity of the test species, evaluate comparability between laboratory test results, and classify possible variability sources such as healthy organisms, organism batches, laboratory water, and food quality changes. For the continuous assessment of sensitivity and accuracy of laboratory toxicity tests, control charts are often used. The cumulative sum (CUSUM) charting structure is a popular monitoring tool that is often used due to its outstanding ability in recognizing small‐to‐moderate unusual changes in process parameters. The CUSUM control charts have now been widely accepted/applied for monitoring the process dispersion under the assumption that the process does not have outliers. In practice, most of the processes have outliers. Therefore, the use of the interquartile range (IQR) as a plotting statistic is an attractive choice for dispersion monitoring. Hence, in this study, the CUSUM control chart with auxiliary information‐based IQR estimators is proposed and employed to monitor dispersion or variation in the variable LC50 by using quantitative structure–activity relationship (QSAR) aquatic toxicity data set. The proposed IQR‐CUSUM charts are compared to its counterparts by using a variety of main output run‐length functions, such as average run length (ARL), standard deviation of run length (SDRL), and extra quadratic loss (EQL), derived from the distribution of run‐length. Further, results revealed that the proposed CUSUM charts have better detection ability as compared to existing counterpart charts, both in theoretical and practical considerations.

show abstract

Prediction of Acute Aquatic Toxicity toward Daphnia Magna by using the GA-kNN Method

Cited by 65 publications

References 55 publications

Validation and extension of a similarity-based approach for prediction of acute aquatic toxicity towards Daphnia magna

Validation and extension of a similarity-based approach for prediction of acute aquatic toxicity towards Daphnia magna

Chemometric Modeling of Daphnia Toxicity

IQR CUSUM charts: An efficient approach for monitoring variations in aquatic toxicity

Contact Info

Product

Resources

About