Considerations and recent advances in QSAR models for cytochrome P450-mediated drug metabolism prediction

Abstract

Quantitative structure-activity relationships (QSAR) methods are urgently needed for predicting ADME/T (absorption, distribution, metabolism, excretion and toxicity) properties to select lead compounds for optimization at the early stage of drug discovery, and to screen drug candidates for clinical trials. Use of suitable QSAR models ultimately results in lesser time-cost and lower attrition rate during drug discovery and development. In the case of ADME/T parameters, drug metabolism is a key determinant of metabolic stability, drug-drug interactions, and drug toxicity. QSAR models for predicting drug metabolism have undergone significant advances recently. However, most of the models used lack sufficient interpretability and offer poor predictability for novel drugs. In this review, we describe some considerations to be taken into account by QSAR for modeling drug metabolism, such as the accuracy/consistency of the entire data set, representation and diversity of the training and test sets, and variable selection. We also describe some novel statistical techniques (ensemble methods, multivariate adaptive regression splines and graph machines), which are not yet used frequently to develop QSAR models for drug metabolism. Subsequently, rational recommendations for developing predictable and interpretable QSAR models are made. Finally, the recent advances in QSAR models for cytochrome P450-mediated drug metabolism prediction, including in vivo hepatic clearance, in vitro metabolic stability, inhibitors and substrates of cytochrome P450 families, are briefly summarized.

Computational methods for inference and prediction of novel drug-drug interactions

“…In 2008, Li et al reviewed the use of QSAR models to predict P450 metabolism, clearance and metabolic stability [7]. They also made rational recommendations for developing predictable and interpretable QSAR models.…”

“…They also made rational recommendations for developing predictable and interpretable QSAR models. These recommendations include the importance of appropriate division of the training and test sets, choosing applicable cut points, and the construction of rationally designed molecular descriptors [7]. Type3 Papers in this type prefer to focus on the specific means or algorithm to build models.…”

ADMEProfiling in Drug Discovery and Development: An Overview

“…In silico intestinal device (ISID) model has also been developed to model absorption for compounds that are dual substrates of CYP and Pgp [69]. QSAR models to predict drug metabolism have been extensively developed using various statistical techniques [MLR (Multiple Linear Regression), PLS (Partial Least Square), NBC (Naïve Bayes Classifier), k-NN, SOM (Self Organizing Map), RP (Recursive Partitioning), ANN (Artificial Neural Networks), and SVM (Support Vector Machines)] [70][71][72]. A recent MLR approach for human hepatic clearance prediction using in vitro experimental data and six molecular descriptors (MW, number of total atoms, number of aromatic rings, number of single bonds, TPSA, and SKlogP) yielded ∼85% success [70].…”

Combination of kernel PCA and linear support vector machine for modeling a nonlinear relationship between bioactivity and molecular descriptors

“…Structure-activity relationship (SAR), a very important area of chemometrics in the modern pharmaceutical industry, is urgently needed for predicting absorption, distribution, metabolism, excretion, toxicity (ADMET) properties to select lead compounds for optimization at the early stage of drug discovery and to screen drug candidates for clinical trials [1]. Much effort in recent SAR studies has been focused on predicting pharmacokinetic and toxicological properties that are collectively referred to as ADMET of compounds.…”