Designing Antibacterial Compounds through a Topological Substructural Approach

Abstract

A novel application of TOPological Substructural MOlecular DEsign (TOPS-MODE) was carried out in antibacterial drugs using computer-aided molecular design. Two series of compounds, one containing antibacterial and the other containing non-antibacterial compounds, were processed by a k-means cluster analysis in order to design training and predicting series. All clusters had a p-level < 0.005. Afterward, a linear classification function has been derived toward discrimination between antibacterial and non-antibacterial compounds. The model correctly classifies 94% of active and 86% of inactive compounds in the training series. More specifically, the model showed a global good classification of 91%, i.e., 263 cases out of 289. In predicting series, the model has shown overall predictabilities of 91 and 83% for active and inactive compounds, respectively. Thereby, the model has a global percentage of good classification of 89%. The TOPS-MODE approach, also, similarly compares with respect to one of the most useful models for antimicrobials selection reported to date.

Inductive QSAR Descriptors. Distinguishing Compounds with Antibacterial Activity by Artificial Neural Networks

“…The latest series of QSAR works report effective separation of bioactive substances from the non-active chemicals by applying the methods of Support Vector Machines (SVM) [17,18], probability-based classification [19], the Artificial Neural Networks (ANN) [20][21][22] and the Bayesian Neural Networks (BNN) [23,24] among others. Several groups used datasets of antibacterial compounds to build the binary classifiers of general antibacterial activity (antibiotic-likeness models) utilizing the ANN algorithm [25][26][27], linear discriminant analysis (LDA) [28,29], binary logistic regression [29] or k-means cluster method [30]. Thus, in the study [31] the LDA has been used to relate anti-malarial activity of a series of chemical compounds to molecular connectivity QSAR indices.…”

Automatic seizure detection based on star graph topological indices

“…In previous works, several graphical representation have been used (lattice, spiral and star graphs) to characterize a diversity of complex systems such as drugs (Dehmer et al, 2009;Molina et al, 2004), proteins (Chen et al, 2010a(Chen et al, , 2010bHu et al, 2011;Huang et al, 2011;, nucleic acids (González-Díaz et al, 2005), proteome mass spectra (Cruz-Monteagudo et al, 2008a) or drug action on parasites (Prado-Prado et al, 2008). A graph is the abstract representation of a real complex network and it consists of nodes (vertices) and links between some of them with similar characteristics (Harary, 1969).…”

Factor Analysis Scale of Generalized Amino Acid Information as the Source of a New Set of Descriptors for Elucidating the Structure and Activity Relationships of Cationic Antimicrobial Peptides

“…The k-means cluster analysis may be used in training and predicting series design [24,37,38]. In this work, we first used k-means cluster analysis to divide the whole data into two groups.…”

“…Modeling technique and methodology for QSAR involve machine learning and statistical techniques such as artificial neural networks [19,20], linear discriminant analysis [21 -23], binary logistic regression [22], principal component analysis, and k-means cluster method [24]. One of the major challenges in modeling methods is the curse of dimensionality [25], in which the dimension of input descriptors is much higher than the number of observed samples, and many QSAR models are unable to generate statistically significant decision boundaries.…”