DEREK, TOPKAT, MCASE are widely accepted softwares. 'QSAR Toolbox', 'TEST' , 'LAZAR', 'TOXTREE' are freely available. You can check Table 2 in attached manuscript.
When assessing the properties of chemicals, the easiest and most consistent way of applying (Quantitative) Structure-Activity Relationship ([Q]SAR) models is to use ready-made software that implements the models via a user-friendly interface. A wide range of software tools are available for predicting physicochemical properties, toxicological endpoints and other biological effects, as well as fate in the environment and biological organisms. Typically, a given software package predicts multiple properties and endpoints, and some are extensible, allowing the user to develop new models or include new knowledge. In addition to (Q)SAR models and rulebases that are incorporated in software tools, there is a growing scientific literature which reports thousands of (Q)SARs. In this report, we give an overview of the software packages that are commonly used in the assessment of chemical toxicity. These software packages are potentially useful in the hazard and risk assessment of chemicals, including for regulatory purposes. However, the applicability of any given software tool needs to be carefully evaluated and documented.
COMPOUND TOXICITY DETECTION SOFTWARE TOOLS | DRUG DISCOVERY DATA ANALYSIS
Determining the toxicity of chemicals is necessary to identify their harmful effects on humans, animals, plants, or the environment. It is also one of the main steps in drug design.
FAF-Drugs / Free ADMET Filteringforum (1)Allows users to detect small molecules. FAF-Drugs is a web application which allows users to filter large compound libraries or determine some
Chapter 5 - In Silico Approaches for Predictive Toxicology
In silico toxicology plays a vital role in the assessment of safety/toxicity of chemicals and the drug development process. Computational approaches continue to increase in capability and applicability to predictive toxicology. These advanced methodology are utilized in various stages of the development of substance by prediction of properties that correlate with toxicity endpoints, structure activity relationship models for new chemical formulations, and building/retrieving information on chemical databases. This chapter covers different aspects of computational approaches that focuses on in silico toxicology, that aims to complement prevailing in vitro/in vivo toxicity tests to predict toxicity and prioritize chemicals/drugs to minimize harmful effects. The state-of-the-art computational approaches used in in silico toxicology are highlighted in this chapter. Special attention has been drawn on the usefulness of quantitative structure activity relationship models for toxicity prediction, descriptor development for predictive toxicology and databases/in silico tools used for toxicity prediction.
Chapter In Silico Approaches for Predictive Toxicology
The Toxicity Estimation Software Tool (TEST) was developed to allow users to easily estimate the toxicity of chemicals using Quantitative Structure Activity Relationships (QSARs) methodologies. QSARs are mathematical models used to predict measures of toxicity from the physical characteristics of the structure of chemicals (known as molecular descriptors). Simple QSAR models calculate the toxicity of chemicals using a simple linear function of molecular descriptors:
Toxicity = ax1 + bx2 + c
Software Disclaimer
TEST estimates the toxicity values and physical properties of organic chemicals based on the molecular structure of the organic chemical entered by the user. The United States Environmental Protection Agency (U.S. EPA) makes no warranty, expressed or implied, as to the merchantability of TEST or its fitness for a particular purpose. Furthermore, the US EPA makes no claims concerning the accuracy of the data provided by TEST or its reliability for any purpose.
where x1 and x2 are the independent descriptor variables and a, b, and c are fitted parameters. The molecular weight and the octanol-water partition coefficient are examples of molecular descriptors. Additional examples are provided in our Molecular Descriptors Guide Version 1.0.2.
TEST allows a user to estimate toxicity without requiring any external programs. Users input a chemical to evaluate by drawing it in an included chemical sketcher window, entering a structure text file, or importing it from an included database of structures. Once entered, the toxicity is estimated using one of several advanced QSAR methodologies. The required molecular descriptors are calculated within TEST.
QSAR Methodologies
Several QSAR methodologies have been developed:
Hierarchical method – The toxicity for a given query compound is estimated using the weighted average of the predictions from several different models. The different models are obtained by using Ward’s method to divide the training set into a series of structurally similar clusters. A genetic algorithm-based technique is used to generate models for each cluster. The models are generated prior to runtime.
FDA method – The prediction for each test chemical is made using a new model that is fit to the chemicals that are most similar to the test compound. Each model is generated at runtime.
Single-model method – Predictions are made using a multilinear regression model that is fit to the training set (using molecular descriptors as independent variables) using a genetic algorithm-based approach. The regression model is generated prior to runtime.
Group contribution method – Predictions are made using a multilinear regression model that is fit to the training set (using molecular fragment counts as independent variables). The regression model is generated prior to runtime.
Nearest neighbor method – The predicted toxicity is estimated by taking an average of the three chemicals in the training set that are most similar to the test chemical.
Consensus method – The predicted toxicity is estimated by taking an average of the predicted toxicities from each of the above QSAR methodologies.
Mode of action method - The predicted toxicity is calculated using a two-step process: (1) linear discriminant models are used to predict the aquatic toxicity mode of action and (2) the quantitative toxicity is predicted using the multiple linear regression model developed for that mode of action.
VirtualToxLab Allows to rationalize a prediction at the molecular level by analyzing the binding mode of the tested compound towards all 16 target proteins in real-time 3D/4D. The VirtualToxLab is an in silico tool for predicting the toxic potential (endocrine and metabolic disruption, some aspects of carcinogenicity and cardiotoxicity) of drugs, chemicals and natural products. It simulates and quantifies their interactions towards a series of proteins known to trigger adverse effects using automated, flexible docking combined with multi-dimensional QSAR (mQSAR). Currently, the VirtualToxLab comprises 16 models of proteins known or suspected to trigger adverse effects: the androgen, aryl hydrocarbon, estrogen α, estrogen β, glucocorticoid, hERG, liver X, mineralocorticoid, progesterone, thyroid α, thyroid β and peroxisome proliferator-activated receptor γ as well as the enzymes CYP450 1A2, 2C9, 2D6 and 3A4.
PASS / Prediction of Activity Spectra for Substances Predicts many kinds of biological activity for compounds from different chemical series based on their 2D structural formulas. PASS finds new targets mechanisms for some ligands. It can reveal new ligands for some biological targets. The tool can be used to analyze the occurrence, in a database, of compounds predicted to be active for a well-defined set of PASS activities.
📷
Derek Nexus Gives accurate toxicity predictions quickly. Derek Nexus is a knowledge-based expert systems that predicts the toxicity and metabolism of a chemical, respectively. It offers an effective mechanism for the sharing of data and knowledge on chemical toxicity and metabolism. It also provides a more direct assessment of predictive performance, avoiding the inherent difficulties of reference to published studies, by allowing the user to access information directly on predictive performance for an alert within the version of the software.
MC-3DQSAR / RNA 3D structure QSAR
A computational method to determine the vital elements of target RNA structure from mutagenesis and available high-resolution data.
Sarah Nexus Gives accurate mutagenicity predictions quickly. Sarah Nexus is a statistical-based system that uses a unique machine-learning methodology to build a statistical model for Ames mutagenicity. This novel system for the fragmentation of query structures enables greater transparency and aids interpretability of predictions in order to assist the user.
GUSAR / General Unrestricted Structure-Activity Relationships Allows pharmacophore identification. GUSAR is based on quantitative neighborhoods of atoms (QNA), multilevel neighborhoods of atoms (MNA) descriptors, and self-consistent regression (SCR) algorithm. It offers idea about the contribution of individual atom in deciding biological activity. The tool permits to point out favorable and unfavorable atoms by suitably coloring the atoms.
ChemoPy / Chemoinformatics in python
An open-source python package for calculating the commonly used structural and physicochemical features. ChemoPy computes 16 drug feature groups composed of 19 descriptors that include 1135 descriptor values. In addition, it provides seven types of molecular fingerprint systems for drug molecules, including topological fingerprints, electro-topological state (E-state) fingerprints, MACCS keys, FP4 keys, atom pairs fingerprints, topological torsion fingerprints and Morgan/circular fingerprints. By applying a semi-empirical quantum chemistry program MOPAC, ChemoPy can also compute a large number of 3D molecular descriptors conveniently.
ROSC-Pred Allows prediction of general rodent carcinogenicity. ROSC-Pred is based on the PASS (Prediction of Activity Spectra for Substances) algorithm. It employs structural formula of organic compounds to proceed. PASS can be used to determine the biological activity spectra for drug-like organic molecules on the basis of their structural formulae. This tool employs the Structure-Activity Relationships database such as set of bases.
eMolTox Predicts different kinds of toxic endpoints from toxicology related in vitro/in vivo experimental data and analyses toxic substructure. eMolTox is a web server that permits users to obtain information
This tutorial demonstrates the OpenTox toxicity prediction application ToxPredict, which accepts chemical structures and names as input and automatically generates a toxicity report based on various precalibrated toxicity models.
Download a Tutorial handout Download a data file to be used in exercise 3 (right click --> Save (Download) Link (Linked File) As)
Introduction
ToxPredict estimates the chemical hazard of chemical structures. It relies on OpenTox API-v1.1 compliant RESTful webservices. Users can either search the OpenTox prototype database, which includes currently quality labelled data for ~150,000 chemicals, grouped in number of datasets, or upload their own chemical structure data. ToxPredict provides access to 16 ready to use models, addressing 14 different endpoints (and growing).
ToxPredict uses the following OpenTox webservices: Compound, Feature, Dataset, Algorithm, Model , Task , Ontology, and Validation.
A Model, available via the OpenTox API, can be easily integrated in ToxPredict by just publishing its Web address.
EVALUATION FORM
When you are done with this tutorial, please fill out the Evaluation Form. Exercise 1 Search for a structure by chemical name, run models and obtain predictions, explore toxicity data for the selected compound, and browse a dataset with toxicity data. Find the tutorial instructions here.
EXERCISE 2
Select subsets of models to be applied, draw a structure, search for similar compounds, view the results and obtain predictions, explore toxicity data for the selected compounds, run model predictions in batch mode, and browse the results. Find the tutorial instructions here.
EXERCISE 3
Upload new chemical structure, run models and obtain predictions, and explore the “My uploads” page. Find the tutorial instructions here.
Just wanted to check whether the QSAR module, Protox II applies 2 methods (expert rule based and statistical based) in order to make sure we are in compliance with ICH M7 Guidance to evaluate the chemcial structure for mutagenicity.
Are there any other freely available software or webservers that take input multiple smiles ID at the same time and predicts their toxicity risk? Similar to what SWISS ADME does...
You can check alvaRunner (https://www.alvascience.com/alvarunner/), it is free for academic purposes and can load multiple SMILES/SDF. A set of models is available at: https://www.alvascience.com/model-studies/