▌QSAR modeling

(quantitative) Structure activity relationship. The (quantitative) model relationship between material effects and molecular descriptors can qualitatively/quantitatively predict the physical and chemical, toxicology, ecotoxicology and environmental behavior of substances based on the structure of substances. (Q)SAR was first applied to drug and pesticide research and development. In recent years,(Q)SAR has been applied to substance toxicity prediction and experimental results prediction.

▌QSAR modeling method

QSAR model construction usually includes the following steps:

step1 Clearly calculated activity endpoint

step2 Data collection

step3 Compute molecular structure descriptors

step4 Model construction algorithm

step5 Model Verification and Evaluation

step6 Model application domain characterization

step7 Mechanism explanation

▌OECD Guidelines for QSAR Model Construction and Application

In order to standardize the application of QSAR technology, OECD issued guidelines on the construction and verification of QSAR models in 2007. The guidelines stipulate that QSAR models used for chemical risk assessment and management should meet the following five requirements: ① The activity of the predicted compounds must have a clear definition; (2) Use clear and clear algorithms; (3) Define the application domain of the model; (4) The model has appropriate fit, stability and predictive ability; (5) It is best to be able to explain the mechanism.

▌Our service

Customized QSAR model based on OECD guidelines.

▌Molecular docking

Molecular simulation generally refers to the technology based on computer simulation of molecular properties and reactions. The results of molecular simulation can explain experimental phenomena, reveal the mechanism of microscopic processes, and can also assist experimental design and predict the properties of molecules. With the gradual improvement of computational chemistry theory and the rapid development of high-performance computing, molecular simulation will play an increasingly important role in environmental computational chemistry and computational toxicology. The simulation of chemical substances can fall on the micro scale, mesoscopic scale and macro scale, and the molecular simulation method is mainly on the micro scale. The three common molecular simulation methods include quantum mechanics (QM) method, molecular mechanics (MM) method and the coupling of the two (QM/MM) method.

▌Quantum chemical calculation

Quantum chemistry adopts the basic principles of quantum mechanics to study the electronic structure, energy, molecular interaction, chemical reaction and other theoretical disciplines of target systems such as atoms and molecules. The core of quantum chemical calculations is to solve the Schrödinger equation to obtain the molecular orbital energy level, energy and other information of the system. This information provides a reliable theoretical basis for judging the active sites of intermolecular interactions, predicting chemical reaction products, designing new drugs and materials, and can also indicate the feasibility direction for experimental research. Common theoretical methods include: valence bond (valence bond, VB) theory, molecular orbital (molecular orbital, MO) theory such as semi-empirical molecular orbital theory, AB initio method, density functional theory, etc. In recent years, the application of quantum chemical calculations in the field of environmental chemistry has enabled people to have a deeper understanding of the migration, transformation and return of organic pollutants in the environment, and promoted the development of science.

▌Software and database services

◆ QSAR toolbox

Is jointly developed by OECD and ECHA and integrates a large number of databases, mechanism classification and prediction models. It can be used for data query, similarity search, compound classification, read-across prediction, QSAR prediction and QSAR model construction.

◆ Toxtree

Is developed by EU JRC, with friendly interface and simple operation. The classification and prediction of substances are mainly based on structural warnings, and are often used for predictive analysis of genotoxicity, skin sensitization and aquatic ecotoxicity.