A genetic algorithm for structure-based de novo design

Genetic algorithms have properties which make them attractive in de novo drug design. Like other de novo design programs, genetic algorithms require a method to reduce the enormous search space of possible compounds. Most often this is done using information from known ligands. We have developed the ADAPT program, a genetic algorithm which uses molecular interactions evaluated with docking calculations as a fitness function to reduce the search space. ADAPT does not require information about known ligands. The program takes an initial set of compounds and iteratively builds new compounds based on the fitness scores of the previous set of compounds. We describe the particulars of the ADAPT algorithm and its application to three well-studied target systems. We also show that the strategies of enhanced local sampling and re-introducing diversity to the compound population during the design cycle provide better results than conventional genetic algorithm protocols.     

Finding transition states using reduced potential-energy surfaces

 We propose a method to locate saddle points that is based on the interplay between the driving coordinate and the restricted quasi-Newton algorithm. The method locates the transition state using a reduced potential-energy surface. The reduced potential-energy surface is characterized by the set of driving coordinates. The proposed algorithm starts at a point on the surface that is slightly perturbed from either reactant or product and, in principle, converges to the transition state. Finally we give a special type of update Hessian matrix formula that should be applied in optimizations carried out on reduced potential-energy surfaces. Received: 29 September 2000 / Accepted: 3 January 2001 / Published online: 3 April 2001     

软件技术在近红外光谱定量分析中的应用

为了更好地运用化学计量学方法提高近红外光谱定量分析的质量,全面研究了相关算法技术与软件工具的应用。根据近红外光谱定量分析的一般流程,依次探讨了光谱数据预处理、异常样本检测、样本集划分、波长选择、模型建立及其参数优化的常用算法,并给出了其中各种主流算法的基本步骤;然后介绍了OMNIC、OPUS、Origin、The Unscrambler X和MATLAB等软件工具的功能特点和主要用途,可为相关技术研究和分析检测工作提供参考。     

Rapid Diagnosis of Type II Diabetes Using Fourier Transform Mid-Infrared Attenuated Total Reflection Spectroscopy Combined with Support Vector Machine

Type II diabetes was diagnosed by Fourier transform mid-infrared (FTMIR) attenuated total reflection (ATR) spectroscopy in combination with support vector machine (SVM). Spectra of serum samples from 65 patients with clinical confirmed type II diabetes mellitus and 55 healthy volunteers were acquired using ATR-FTMIR and were first pretreated by three pretreatments (Savitzky–Golay smoothing, multiple scattering correction, and wavelet transforms algorithms) to reduce the interfering information before establishing the SVM models. The parameters of SVM (penalty factor C and kernel function parameter gamma) were optimized to improve the generalization abilities of the models. A grid search method (GS), genetic algorithm (GA), and particle swarm optimization (PSO) algorithm, were used to find out the optimal parameter values. The results showed that the maximum accuracies were 95.74, 97.87, and 89.36% for the optimized GS, GA, and PSO algorithms. The maximum sensitivities were 96, 100, and 92, and the maximum specificity were 95.45, 95.45, and 86.36%, respectively. The results indicated that the accuracy of type II diabetes was improved using the GS, GA, and PSO algorithms for optimizing the SVM parameters. The GA was found to be slightly better than the GS and PSO. The results of the experiment confirmed that the combination of the ATR-FTMIR spectroscopy and SVM was able to rapidly and accurately diagnose type II diabetes without reagents.     

Molecular Similarity Based on Super-Ball Algorithm

IntroductionIn recent years, the studies about mo1ecular similarity has become an active research fieldand there has been an upsurge of interest in the use of the molecular similarity method fordrug design, the selection of analogs for chemicals, and the estimation of properties['--'J.There is a long-standing axiom in the molecular similarity method, depending on whichmolecules which are structura1ly similar, by specified criteria, tend to have similar chemicaland biological activities. In mol…     

Multicanonical parallel simulations of proteins with continuous potentials

The determination of the three‐dimensional (3D) structure of a protein or peptide is a very important research problem in biological and medical sciences. Anfinsen's experiments (Science 1973, 181, 223) on renaturation of denatured proteins have shown that the native 3D structure of a (small) protein at low (room) temperatures is uniquely determined by its amino acid sequence, which suggests that it might be possible to determine the 3D structure of a protein from its amino acid sequence by pure computations. As a step toward that goal, in this article we present a simple approach for parallelization of multicanonical Monte Carlo simulations of proteins with continuous potentials. Our method is based on the parallel calculation of the protein energy function. The algorithm is tested by simulated annealing and multicanonical simulations of two small peptides, and known results are reproduced accurately. An acceptable degree of parallelization can be achieved in the simulation of Protein L using up to 30 PCs. © 2001 John Wiley & Sons, Inc. J Comput Chem 22: 1287–1296, 2001     

Variable selection for multivariate calibration using a genetic algorithm: prediction of additive concentrations in polymer films from Fourier transform-infrared spectral data

Variable selection using a genetic algorithm is combined with partial least squares (PLS) for the prediction of additive concentrations in polymer films using Fourier transform-infrared (FT-IR) spectral data. An approach using an iterative application of the genetic algorithm is proposed. This approach allows for all variables to be considered and at the same time minimizes the risk of overfitting. We demonstrate that the variables selected by the genetic algorithm are consistent with expert knowledge. This very exciting result is a convincing application that the algorithm can select correct variables in an automated fashion.     

A genetic algorithm for the automated generation of small organic molecules: Drug design using an evolutionary algorithm

Rational drug design involves finding solutions to large combinatorial problems for which an exhaustive search is impractical. Genetic algorithms provide a novel tool for the investigation of such problems. These are a class of algorithms that mimic some of the major characteristics of Darwinian evolution. LEA has been designed in order to conceive novel small organic molecules which satisfy quantitative structure-activity relationship based rules (fitness). The fitness consists of a sum of constraints that are range properties. The algorithm takes an initial set of fragments and iteratively improves them by means of crossover and mutation operators that are related to those involved in Darwinian evolution. The basis of the algorithm, its implementation and parameterization, are described together with an application in de novo molecular design of new retinoids. The results may be promising for chemical synthesis and show that this tool may find extensive applications in de novo drug design projects.     

Combination of RSM and NSGA-II algorithm for optimization and prediction of thermal conductivity and viscosity of bioglycol/water mixture containing SiO2 nanoparticles

The fluids containing nanoparticles have enhanced thermo-physical characteristics in comparison with conventional fluids without nanoparticles. Thermal conductivity and viscosity are thermo-physical properties that strongly determine heat transfer and momentum. In this study, the response surface method was firstly used to derive an equation for the thermal conductivity and another one for the viscosity of bioglycol/water mixture (20:80) containing silicon dioxide nanoparticles as a function of temperature as well as the volume fraction of silicon dioxide. Then, NSGA-II algorithm was used for the optimization and maximizing thermal conductivity and minimizing the nanofluid viscosity. Different fronts were implemented and 20th iteration number was selected as Pareto front. The highest thermal conductivity (0.576 W/m.K) and the lowest viscosity (0.61 mPa.s) were obtained at temperature on volume concentration of (80 °C and 2%) and (80 °C without nanoparticle) respectively. It was concluded that the optimum thermal conductivity and viscosity of nanofluid could be obtained at maximum temperature (80 °C) or a temperature close to this temperature. An increase in the volume fraction of silicon dioxide led to the enhancement of thermal conductivity but the solution viscosity was also increased. Therefore, the optimum point should be selected based on the system requirement.