Application of partial least squares and radial basis function neural networks in multivariate imaging analysis-quantitative structure activity relationship: Study of cyclin dependent kinase 4 inhibitors

Isonicotinic Acid Hydrazide Derivatives: Synthesis, Antimycobacterial, Antiviral, Antimicrobial Activity and QSAR Studies

“…Isonicotinic acid (4-hydroxy-3,5-dimethoxy-benzylidene)-hydrazide (13) was dissolved in dichloromethane (0.001 mol) and a solution of benzoyl chloride (0.001 mol), obtained by the reaction of benzoic acid with thionyl chloride, in dichloromethane was added drop wise and kept for stirring on a magnetic stirrer till the evaporation of dichloromethane. The product obtained was checked for purity by TLC and recrystallized from ethanol.…”

“…The purpose in quantitative structure activity relationship (QSAR) methodology is to construct a relationship between physicochemical properties as independent and bioactivity of ligands as a dependent variable. Physicochemical properties could be obtained as descriptors and then in silico methods are applied to manipulate the information, remove noise and derive useful information [13].…”

Compressed images for affinity prediction-2 (CIFAP-2): an improved machine learning methodology on protein–ligand interactions based on a study on caspase 3 inhibitors

“…In other words, docking energy functions mostly employ electrostatic and Van der Waals energy terms determined in space, but not in aqueous media. Intelligent computational methods have recently become popular in drug design [2][3][4][5][6][7][8][9][10] . It is possible to search for or predict specific properties of new drugs by using information from known ligand-receptor interactions in association with bioinformatics and machine learning methods 11 .…”

Application of unfolded principal component analysis–radial basis function neural network for determination of celecoxib in human serum by three-dimensional excitation–emission matrix fluorescence spectroscopy

“…The radial basis function (RBF) neural network (RBFNN) came into fashion since Broomhead and Lowew's seminal paper in 1988 [17,18]. RBF neural networks are a type of ANNs for application to problems of supervised learning, such as regression [19,20], pattern recognition [21], etc. RBF neural networks are local approximation networks.…”

“…On the other hand, the network tends to overfit the training data, if too many hidden layer units are used. Based on our previous experiences about RBFNN[19,20], 6 hidden nodes and the spread of 1.0 were selected for an initial network.Using 8 PCs, the RBNN was trained and the corresponding RMSECV for the calibration set was calculated. The number of hidden nodes was changed from n h =1 to n h =15 and SPREAD from 0.01 to 2 in increments of 0.01.…”