Evolutionary Computation and QSAR Research

Abstract

The successful high throughput screening of molecule libraries for a specific biological property is one of the main improvements in drug discovery. The virtual molecular filtering and screening relies greatly on quantitative structure-activity relationship (QSAR) analysis, a mathematical model that correlates the activity of a molecule with molecular descriptors. QSAR models have the potential to reduce the costly failure of drug candidates in advanced (clinical) stages by filtering combinatorial libraries, eliminating candidates with a predicted toxic effect and poor pharmacokinetic profiles, and reducing the number of experiments. To obtain a predictive and reliable QSAR model, scientists use methods from various fields such as molecular modeling, pattern recognition, machine learning or artificial intelligence. QSAR modeling relies on three main steps: molecular structure codification into molecular descriptors, selection of relevant variables in the context of the analyzed activity, and search of the optimal mathematical model that correlates the molecular descriptors with a specific activity. Since a variety of techniques from statistics and artificial intelligence can aid variable selection and model building steps, this review focuses on the evolutionary computation methods supporting these tasks. Thus, this review explains the basic of the genetic algorithms and genetic programming as evolutionary computation approaches, the selection methods for high-dimensional data in QSAR, the methods to build QSAR models, the current evolutionary feature selection methods and applications in QSAR and the future trend on the joint or multi-task feature selection methods.

Development of HIV-1 integrase inhibitors: recent molecular modeling perspectives

“…It builds a direct quantitative relation between antiviral activities and physicochemical parameters of a series of IN inhibitors, which decreases the number of compounds needing to be synthesized by facilitating the selection of the most promising candidates [27]. The fundamental underlying principle is that any change in a chemical structure results in a positive or negative change in the bioactivity despite the existence of 'activity cliffs' [28] (i.e., pairs of structurally similar compounds that display a large difference in potency against a given target).…”

Xenobiotic Metabolism Activation as a Biomarker of Cigarette Smoke Exposure Response

“…The descriptors for 2D-QSAR can be comprised of ordinary descriptors (e.g., molecular weights, number of atoms, type of bonds, and number of aromatic rings); electrostatic descriptors (e.g., atomic, quantum-chemical, and polarizability); and topological descriptors (e.g., molecular surface, and topological pharmacophores) [ 19 , 23 ]. For 3D-QSAR, the descriptors can be alignment-dependent or alignment-independent [ 23 ]. Alignmentdependent descriptors include the comparative molecular fi eld analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) [ 23 ].…”

“…For 3D-QSAR, the descriptors can be alignment-dependent or alignment-independent [ 23 ]. Alignmentdependent descriptors include the comparative molecular fi eld analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) [ 23 ]. Furthermore, alignment-independent descriptors are invariable to molecular rotation and transformation in space, such as Comparative Molecular Moment Analysis (CoMMA), Weighted Holistic Invariant Molecular (WHIM), Detailed information on the number of ligands for training and test datasets were not stated in the original paper Molecular Surface WHIM, VolSuf, and Grid-Independent Descriptors (GRIND) [ 19 , 23 ].…”

“…The descriptors will be further selected based on their contribution to the activity prediction [ 19 , 23 ]; many techniques are available for this selection step. The selected techniques should be applicable for large datasets containing structurally diverse compounds and should optimize a quick and automatic selection [ 23 ]. The selection methods can generally be divided into two major categories: the fi ltering and the wrapper methods [ 19 , 23 ].…”

“…Although linearity relationship might exist in the data, other possible nonlinearity association should not be disregarded; especially if, a priori knowledge is not available for the set of compounds being tested [ 24 ]. In such situations, nonlinear approaches would be more suitable, such as Bayes classifi er, machine learning k-Nearest neighbor, Artifi cial Neural Network (ANN), Decision Trees (DT), Support Vector Machines (SVM), Evolutionaly Algorithms (EA), and Ensemble methods [ 23 ].…”