The performance, reliability and potential application of in silico models for predicting the acute oral toxicity of pharmaceutical compounds

Graham, Jessica C.; Rodas, Maryann; Hillegass, Jedd M.; Schulze, Gene E.

doi:10.1016/j.yrtph.2020.104816

Cited by 18 publications

(12 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, alternative methods have been explored in terms of costs of experiments and ethics. Among them, in silico prediction is attracting attention, and multiple predictive models based on QSAR analysis have been reported. , In particular, the collaborative acute toxicity modeling suite (CATMoS) was used to construct a large-scale predictive model for LD 50 . In the present investigation, we focused on a binary model with the criterion of 50 mg/kg using data sets in the CATMoS in which structural information of compounds was included; predictive models were constructed using the same data set (training set and test set).…”

Section: Discussionmentioning

confidence: 99%

Predictive Models Based on Molecular Images and Molecular Descriptors for Drug Screening

Mamada,

Takahashi,

Ogino

et al. 2023

ACS Omega

View full text Add to dashboard Cite

Various toxicity and pharmacokinetic evaluations as screening experiments are needed at the drug discovery stage. Currently, to reduce the use of animal experiments and developmental expenses, the development of high-performance predictive models based on quantitative structure–activity relationship analysis is desired. From these evaluation targets, we selected 50% lethal dose (LD50), blood–brain barrier penetration (BBBP), and the clearance (CL) pathway for this investigation and constructed predictive models for each target using 636–11,886 compounds. First, we constructed predictive models using the DeepSnap-deep learning (DL) method and images of compounds as features. The calculated area under the curve (AUC) and balanced accuracy (BAC) were, respectively, 0.887 and 0.818 for LD50, 0.893 and 0.824 for BBBP, and 0.883 and 0.763 for the CL pathway. Next, molecular descriptors (MDs) of compounds were calculated using Molecular Operating Environment, alvaDesc, and ADMET Predictor to construct predictive models using the MD-based method. Using these MDs, we constructed predictive models using DataRobot. The calculated AUC and BAC were, respectively, 0.931 and 0.805 for LD50, 0.919 and 0.849 for BBBP, and 0.900 and 0.807 for the CL pathway. In this investigation, we constructed predictive models combining the DeepSnap-DL and MD-based methods. In ensemble models using the mean predictive probability of the DeepSnap-DL and MD-based methods, the calculated AUC and BAC were, respectively, 0.942 and 0.842 for LD50, 0.936 and 0.853 for BBBP, and 0.908 and 0.832 for the CL pathway, with improved predictive performance observed for all variables compared with either single method alone. Moreover, in consensus models that adopted only compounds for which the results of the two methods agreed, the calculated BAC for LD50, BBBP, and the CL pathway were 0.916, 0.918, and 0.847, respectively, indicating higher predictive performance than the ensemble models for all three variables. The predictive models combining the DeepSnap-DL and MD-based methods displayed high predictive performance for LD50, BBBP, and the CL pathway. Therefore, the application of this approach to prediction targets in various drug discovery screenings is expected to accelerate drug discovery.

show abstract

Section: Discussionmentioning

confidence: 99%

Predictive Models Based on Molecular Images and Molecular Descriptors for Drug Screening

Mamada,

Takahashi,

Ogino

et al. 2023

ACS Omega

View full text Add to dashboard Cite

show abstract

“…The use of available data, advancements of new techniques, and the search for alternatives for animal experimentation have created more interest in computational modeling. Although in silico technology cannot eliminate the use of laboratory animals, it has the potential to reduce it (Graham et al., 2021). Eight studies used molecular docking to understand the toxicity of artificial azo dyes.…”

Section: Risk Assessment Of Artificial Azo Dyesmentioning

confidence: 99%

Risk assessment of azo dyes as food additives: Revision and discussion of data gaps toward their improvement

Ramos-Souza

Bandoni

Arisseto-Bragotto

et al. 2022

Comp Rev Food Sci Food Safe

View full text Add to dashboard Cite

The food industry uses dyes mainly to overcome color loss during the processing and storage of products, with the azo dyes currently being the most employed. Studies on the safety of using these dyes in foods started in the 1950s and have indicated the potential for concern. This review discusses the risk assessment of food intake containing artificial azo dyes. There are case reports and, subsequently, double‐blind placebo‐controlled trials in some individuals who may experience adverse effects from the intake of azo dyes, but it is unclear whether these adverse effects are restricted to specific populations or more generalized. In view of this, different toxicological endpoints are evaluated to verify toxic effects in in vitro and in vivo models and to establish the no observed adverse effect level. Exposure estimation studies have shown that human exposure to azo dyes via oral intake is mainly below the acceptable daily intake established by advisory bodies. However, most countries do not have studies that estimate the oral intake of azo dyes. In this case, local food diversity and racial–ethnic specificities are not considered when stating the exposure estimate is below the acceptable daily intake for the human population and thus may not represent actual intake. Concerning the scenario established above, this review discusses the most critical gaps to be overcome to contribute to the direction of future studies and the development of more effective public policies concerning the safety of the intake of artificial azo dyes.

show abstract

“…26,27 The CATMoS models have also been tested in the pharmaceutical industry to assess 371 Bristol Myers Squibb compounds separating molecules with undesirable LD 50 values (LD 50 > 300 mg/kg) from those with low acute oral toxicity (LD 50 > 2000 mg/kg). 30 Our group has participated in this collaboration and initially used the CATMoS dataset for generating binary Bayesian classification models 31 with our first version of our in-house software called Assay Central. 32−34 We generated classification models as most of the literature provided an upper limit for the dose as 2000 or 5000 mg/kg.…”

Section: ■ Introductionmentioning

confidence: 99%

“…Numerous computational or in silico models have been developed to predict acute toxicity using the previously generated in vivo data, and these are considered acceptable by many researchers and regulatory agencies. ,− A recently curated dataset for rat acute oral toxicity from NTP Interagency Center for the Evaluation of Alternative Toxicological Methods (NICEATM) and EPA National Center for Computational Toxicology (NCCT) , has been used to generate machine learning models by various groups with different machine learning algorithms and molecular descriptors. , This facilitated the Collaborative Acute Toxicity Modeling Suite (CATMoS) representing the generated consensus predictions from each method used by different groups, which leverages the collective strengths of each individual model used. CATMoS demonstrated high performance in terms of accuracy and robustness compared with in vivo results. , The CATMoS models have also been tested in the pharmaceutical industry to assess 371 Bristol Myers Squibb compounds separating molecules with undesirable LD 50 values (LD 50 > 300 mg/kg) from those with low acute oral toxicity (LD 50 > 2000 mg/kg) …”

Section: Introductionmentioning

confidence: 99%

Comparing LD₅₀/LC₅₀ Machine Learning Models for Multiple Species

Lane

Harris

Urbina

et al. 2023

ACS Chem. Health Saf.

View full text Add to dashboard Cite

The lethal dose or concentration which kills 50% of the animals (LD50 or LC50) is an important parameter for scientists to understand the toxicity of chemicals in different scenarios that can be used to make go-no-go decisions, and ultimately assist in the choice of the right personal protective equipment needed for containment. The LD50 assessment process has also required the use of many animals although modern methods have reduced the number of rats needed. Since a compound is usually considered highly toxic when the LD50 is lower than 25 mg/kg, such a classification provides potentially valuable safety information to synthetic chemists and other safety assessment scientists. The need for finding alternative approaches such as computational methods is important to ultimately reduce animal use for this testing further still. We now summarize our efforts to use public data for building in vivo LD50 or LC50 classification and regression machine learning models for various species (rat, mouse, fish, and daphnia) and their fivefold cross-validation statistics with different machine learning algorithms as well as an external curated test set for mouse LD50. These datasets consist of different molecule classes, may cover different activity ranges, and also have a range of dataset sizes. The challenges of using such computational models are that their applicability domain will also need to be understood so that they can be used to make reliable predictions for novel molecules. These machine learning models will also need to be backed up with experimental validation. However, such models could also be used for efforts to bridge gaps in individual toxicity datasets. Making such models available also opens them up to potential misuse or dual use. We will summarize these efforts and propose that they could be used for scoring the millions of commercially available molecules, most of which likely do not have a known LD50 or for that matter any data in vitro or in vivo for toxicity.

show abstract

The performance, reliability and potential application of in silico models for predicting the acute oral toxicity of pharmaceutical compounds

Cited by 18 publications

References 12 publications

Predictive Models Based on Molecular Images and Molecular Descriptors for Drug Screening

Predictive Models Based on Molecular Images and Molecular Descriptors for Drug Screening

Risk assessment of azo dyes as food additives: Revision and discussion of data gaps toward their improvement

Comparing LD₅₀/LC₅₀ Machine Learning Models for Multiple Species

Contact Info

Product

Resources

About

The performance, reliability and potential application of in silico models for predicting the acute oral toxicity of pharmaceutical compounds

Cited by 18 publications

References 12 publications

Predictive Models Based on Molecular Images and Molecular Descriptors for Drug Screening

Predictive Models Based on Molecular Images and Molecular Descriptors for Drug Screening

Risk assessment of azo dyes as food additives: Revision and discussion of data gaps toward their improvement

Comparing LD50/LC50 Machine Learning Models for Multiple Species

Contact Info

Product

Resources

About

Comparing LD₅₀/LC₅₀ Machine Learning Models for Multiple Species