Application of 'inductive' QSAR descriptors for quantification of antibacterial activity of cationic polypeptides

In Silico, in Vitro and in Vivo Study of Substituted Imidazolidinone Sulfonamides as Antibacterial Agents

New substituted imidazolidinone sulfonamides have been developed using a rational drug design strategy. Predictive QSAR models for the search of new antibacterials were created using the OCHEM platform. Regression models were applied to verify a virtual chemical library of new imidazolidinone derivatives designed to have antibacterial activity. A number of substituted imidazolidinone sulfonamides as effective antibacterial agents were identified by QSAR prediction, synthesized and characterized by spectral and elemental, and tested in vitro. Six studied compounds have shown the highest in vitro antibacterial activity against Gram-negative E. coli and Gram-positive S. aureus multidrug-resistant strains. The in vivo acute toxicity of these imidazolidinone sulfonamides based on the LC50 value ranged from 16.01 to 44.35 mg/L (slightly toxic compounds class). The results of molecular docking suggest that the antibacterial mechanism of the compounds can be associated with the inhibition of post-translational modification processes of bacterial peptides and proteins.

Using the local symmetry in amino acids sequences of polypeptides to improve the predictive potential of models of their inhibitor activity

The minimal inhibitory concentrations (pMIC) are a valuable measure of the biological activity of polypeptides. Numerical data on the pMIC are necessary to systematize knowledge on polypeptides' biochemical behaviour. The model of negative decimal logarithm of pMIC of polypeptides in the form of a mathematical function of a sequence of amino acids is suggested. The suggested model is based on the so-called correlation weights of amino acids together with the correlation weights of fragments of local symmetry (FLS). Three kinds of the FLS are considered: (i) three-symbol fragments '…xyx…', (ii) four-symbol fragments '…xyyx…', and (iii) five-symbol fragments '…xyzyx…'. The models built using the Monte Carlo technique improved by applying the index of ideality of correlation (IIC) and correlation intensity index (CII).

Facile and Versatile Surface Functional Polyetheretherketone with Enhanced Bacteriostasis and Osseointegrative Capability for Implant Application

Implant-associated infections and inadequate osseointegration are two challenges of implant materials in orthopedics. In this study, a lithium-ion-loaded (Li+)/mussel-inspired antimicrobial peptide (AMP) designed to improve the osseointegration and inhibit bacterial infections effectively is prepared on a polyetheretherketone (PEEK) biomaterial surface through the combination of hydrothermal treatment and mussel-inspired chemistry. The results illustrate that the multifunctional PEEK material demonstrated a great inhibitory effect on Escherichia coli and Staphylococcus aureus, which was attributed to irreversible bacterial membrane damage. In addition, the multifunctional PEEK can simultaneously upregulate the expression of osteogenesis-associated genes/proteins via the Wnt/β-catenin signaling pathway. Furthermore, an in vivo assay of an infection model revealed that the multifunctional PEEK implants killed bacteria with an efficiency of 95.03%. More importantly, the multifunctional PEEK implants accelerated the implant-bone interface osseointegration compared with pure PEEK implants in the noninfection model. Overall, this work provides a promising strategy for improving orthopedic implant materials with ideal osseointegration and infection prevention simultaneously, which may have broad application clinical prospects.

Deep learning improves antimicrobial peptide recognition

Motivation

Bacterial resistance to antibiotics is a growing concern. Antimicrobial peptides (AMPs), natural components of innate immunity, are popular targets for developing new drugs. Machine learning methods are now commonly adopted by wet-laboratory researchers to screen for promising candidates.

Results

In this work, we utilize deep learning to recognize antimicrobial activity. We propose a neural network model with convolutional and recurrent layers that leverage primary sequence composition. Results show that the proposed model outperforms state-of-the-art classification models on a comprehensive dataset. By utilizing the embedding weights, we also present a reduced-alphabet representation and show that reasonable AMP recognition can be maintained using nine amino acid types.

Availability and implementation

Models and datasets are made freely available through the Antimicrobial Peptide Scanner vr.2 web server at www.ampscanner.com.

Supplementary information

Supplementary data are available at Bioinformatics online.

Improving Recognition of Antimicrobial Peptides and Target Selectivity through Machine Learning and Genetic Programming

Growing bacterial resistance to antibiotics is spurring research on utilizing naturally-occurring antimicrobial peptides (AMPs) as templates for novel drug design. While experimentalists mainly focus on systematic point mutations to measure the effect on antibacterial activity, the computational community seeks to understand what determines such activity in a machine learning setting. The latter seeks to identify the biological signals or features that govern activity. In this paper, we advance research in this direction through a novel method that constructs and selects complex sequence-based features which capture information about distal patterns within a peptide. Comparative analysis with state-of-the-art methods in AMP recognition reveals our method is not only among the top performers, but it also provides transparent summarizations of antibacterial activity at the sequence level. Moreover, this paper demonstrates for the first time the capability not only to recognize that a peptide is an AMP or not but also to predict its target selectivity based on models of activity against only Gram-positive, only Gram-negative, or both types of bacteria. The work described in this paper is a step forward in computational research seeking to facilitate AMP design or modification in the wet laboratory.

PepBio: predicting the bioactivity of host defense peptides

A large-scale QSAR study of host defense peptides sheds light on the origin of their bioactivities (antibacterial, anticancer, antiviral and antifungal).

Discovery of vascular endothelial growth factor receptor tyrosine kinase inhibitors by quantitative structure–activity relationships, molecular dynamics simulation and free energy calculation

Employing the combined strategy to understand the features of KDR–ligands complexes, and provide a basis for rational design of inhibitors.

Identification of Peptide Inhibitors of Enveloped Viruses Using Support Vector Machine

The peptides derived from envelope proteins have been shown to inhibit the protein-protein interactions in the virus membrane fusion process and thus have a great potential to be developed into effective antiviral therapies. There are three types of envelope proteins each exhibiting distinct structure folds. Although the exact fusion mechanism remains elusive, it was suggested that the three classes of viral fusion proteins share a similar mechanism of membrane fusion. The common mechanism of action makes it possible to correlate the properties of self-derived peptide inhibitors with their activities. Here we developed a support vector machine model using sequence-based statistical scores of self-derived peptide inhibitors as input features to correlate with their activities. The model displayed 92% prediction accuracy with the Matthew's correlation coefficient of 0.84, obviously superior to those using physicochemical properties and amino acid decomposition as input. The predictive support vector machine model for self- derived peptides of envelope proteins would be useful in development of antiviral peptide inhibitors targeting the virus fusion process.

Peptides and Peptidomimetics for Antimicrobial Drug Design

The purpose of this paper is to introduce and highlight a few classes of traditional antimicrobial peptides with a focus on structure-activity relationship studies. After first dissecting the important physiochemical properties that influence the antimicrobial and toxic properties of antimicrobial peptides, the contributions of individual amino acids with respect to the peptides antibacterial properties are presented. A brief discussion of the mechanisms of action of different antimicrobials as well as the development of bacterial resistance towards antimicrobial peptides follows. Finally, current efforts on novel design strategies and peptidomimetics are introduced to illustrate the importance of antimicrobial peptide research in the development of future antibiotics.

QSAR modeling: where have you been? Where are you going to?

Quantitative structure-activity relationship modeling is one of the major computational tools employed in medicinal chemistry. However, throughout its entire history it has drawn both praise and criticism concerning its reliability, limitations, successes, and failures. In this paper, we discuss (i) the development and evolution of QSAR; (ii) the current trends, unsolved problems, and pressing challenges; and (iii) several novel and emerging applications of QSAR modeling. Throughout this discussion, we provide guidelines for QSAR development, validation, and application, which are summarized in best practices for building rigorously validated and externally predictive QSAR models. We hope that this Perspective will help communications between computational and experimental chemists toward collaborative development and use of QSAR models. We also believe that the guidelines presented here will help journal editors and reviewers apply more stringent scientific standards to manuscripts reporting new QSAR studies, as well as encourage the use of high quality, validated QSARs for regulatory decision making.

Prediction of linear cationic antimicrobial peptides based on characteristics responsible for their interaction with the membranes

Most available antimicrobial peptides (AMP) prediction methods use common approach for different classes of AMP. Contrary to available approaches, we suggest that a strategy of prediction should be based on the fact that there are several kinds of AMP that vary in mechanisms of action, structure, mode of interaction with membrane, etc. According to our suggestion for each kind of AMP, a particular approach has to be developed in order to get high efficacy. Consequently, in this paper, a particular but the biggest class of AMP, linear cationic antimicrobial peptides (LCAP), has been considered and a newly developed simple method of LCAP prediction described. The aim of this study is the development of a simple method of discrimination of AMP from non-AMP, the efficiency of which will be determined by efficiencies of selected descriptors only and comparison the results of the discrimination procedure with the results obtained by more complicated discriminative methods. As descriptors the physicochemical characteristics responsible for capability of the peptide to interact with an anionic membrane were considered. The following characteristics such as hydrophobicity, amphiphaticity, location of the peptide in relation to membrane, charge density, propensities to disordered structure and aggregation were studied. On the basis of these characteristics, a new simple algorithm of prediction is developed and evaluation of efficacies of the characteristics as descriptors performed. The results show that three descriptors, hydrophobic moment, charge density and location of the peptide along the membranes, can be used as discriminators of LCAPs. For the training set, our method gives the same level of accuracy as more complicated machine learning approaches offered as CAMP database service tools. For the test set accuracy obtained by our method gives even higher value than the one obtained by CAMP prediction tools. The AMP prediction tool based on the considered method is available at http://www.biomedicine.org.ge/dbaasp/.

Class IIa bacteriocins: diversity and new developments

Class IIa bacteriocins are heat-stable, unmodified peptides with a conserved amino acids sequence YGNGV on their N-terminal domains, and have received much attention due to their generally recognized as safe (GRAS) status, their high biological activity, and their excellent heat stability. They are promising and attractive agents that could function as biopreservatives in the food industry. This review summarizes the new developments in the area of class IIa bacteriocins and aims to provide uptodate information that can be used in designing future research.

ClassAMP: a prediction tool for classification of antimicrobial peptides

Antimicrobial peptides (AMPs) are gaining popularity as anti-infective agents. Information on sequence features that contribute to target specificity of AMPs will aid in accelerating drug discovery programs involving them. In this study, an algorithm called ClassAMP using Random Forests (RFs) and Support Vector Machines (SVMs) has been developed to predict the propensity of a protein sequence to have antibacterial, antifungal, or antiviral activity. ClassAMP is available at http://www.bicnirrh.res.in/classamp/.

Predicting antitumor activity of peptides by consensus of regression models trained on a small data sample

Predicting antitumor activity of compounds using regression models trained on a small number of compounds with measured biological activity is an ill-posed inverse problem. Yet, it occurs very often within the academic community. To counteract, up to some extent, overfitting problems caused by a small training data, we propose to use consensus of six regression models for prediction of biological activity of virtual library of compounds. The QSAR descriptors of 22 compounds related to the opioid growth factor (OGF, Tyr-Gly-Gly-Phe-Met) with known antitumor activity were used to train regression models: the feed-forward artificial neural network, the k-nearest neighbor, sparseness constrained linear regression, the linear and nonlinear (with polynomial and Gaussian kernel) support vector machine. Regression models were applied on a virtual library of 429 compounds that resulted in six lists with candidate compounds ranked by predicted antitumor activity. The highly ranked candidate compounds were synthesized, characterized and tested for an antiproliferative activity. Some of prepared peptides showed more pronounced activity compared with the native OGF; however, they were less active than highly ranked compounds selected previously by the radial basis function support vector machine (RBF SVM) regression model. The ill-posedness of the related inverse problem causes unstable behavior of trained regression models on test data. These results point to high complexity of prediction based on the regression models trained on a small data sample.

Quantitative structure-activity relationships and docking studies of calcitonin gene-related peptide antagonists

Defining the role of calcitonin gene-related peptide in migraine pathogenesis could lead to the application of calcitonin gene-related peptide antagonists as novel migraine therapeutics. In this work, quantitative structure-activity relationship modeling of biological activities of a large range of calcitonin gene-related peptide antagonists was performed using a panel of physicochemical descriptors. The computational studies evaluated different variable selection techniques and demonstrated shuffling stepwise multiple linear regression to be superior over genetic algorithm-multiple linear regression. The linear quantitative structure-activity relationship model revealed better statistical parameters of cross-validation in comparison with the non-linear support vector regression technique. Implementing only five peptide descriptors into this linear quantitative structure-activity relationship model resulted in an extremely robust and highly predictive model with calibration, leave-one-out and leave-20-out validation R(2) of 0.9194, 0.9103, and 0.9214, respectively. We performed docking of the most potent calcitonin gene-related peptide antagonists with the calcitonin gene-related peptide receptor and demonstrated that peptide antagonists act by blocking access to the peptide-binding cleft. We also demonstrated the direct contact of residues 28-37 of the calcitonin gene-related peptide antagonists with the receptor. These results are in agreement with the conclusions drawn from the quantitative structure-activity relationship model, indicating that both electrostatic and steric factors should be taken into account when designing novel calcitonin gene-related peptide antagonists.

Connecting peptide physicochemical and antimicrobial properties by a rational prediction model

The increasing rate in antibiotic-resistant bacterial strains has become an imperative health issue. Thus, pharmaceutical industries have focussed their efforts to find new potent, non-toxic compounds to treat bacterial infections. Antimicrobial peptides (AMPs) are promising candidates in the fight against antibiotic-resistant pathogens due to their low toxicity, broad range of activity and unspecific mechanism of action. In this context, bioinformatics' strategies can inspire the design of new peptide leads with enhanced activity. Here, we describe an artificial neural network approach, based on the AMP's physicochemical characteristics, that is able not only to identify active peptides but also to assess its antimicrobial potency. The physicochemical properties considered are directly derived from the peptide sequence and comprise a complete set of parameters that accurately describe AMPs. Most interesting, the results obtained dovetail with a model for the AMP's mechanism of action that takes into account new concepts such as peptide aggregation. Moreover, this classification system displays high accuracy and is well correlated with the experimentally reported data. All together, these results suggest that the physicochemical properties of AMPs determine its action. In addition, we conclude that sequence derived parameters are enough to characterize antimicrobial peptides.

Methods for building quantitative structure-activity relationship (QSAR) descriptors and predictive models for computer-aided design of antimicrobial peptides

Antimicrobial peptides are ubiquitous in nature where they play important roles in host defense and microbial control. More than 1,000 naturally occurring peptides have been described so far and those considered for pharmaceutical development have all been further optimized by rational approaches. In recent years, high-throughput screening assays have been developed to specifically address optimization of AMPs. In addition to these cell-based in vivo systems, a range of computational in silico systems can be applied in order to predict the biological activity of AMPs for specific bacteria. Among them, quantitative structure-activity relationships (QSARs) method, which attempts to correlate chemical structure to biological measurement, has shown promising results in the optimization and discovery of peptide candidates. Therefore, this chapter is devoted to describe the QSAR method and recent progress applied in AMP.

Distinguishing compounds with anticancer activity by ANN using inductive QSAR descriptors

This article describes a method developed for predicting anticancer/non-anticancer drugs using artificial neural network (ANN). The ANN used in this study is a feed-forward neural network with a standard back-propagation training algorithm. Using 30 ‘inductive’ QSAR descriptors alone, we have been able to achieve 84.28% accuracy for correct separation of compounds with- and without anticancer activity. For the complete set of 30 inductive QSAR descriptors, ANN based method reveals a superior model (accuracy = 84.28%, Q_pred = 74.28%, sensitivity = 0.9285, specificity = 0.7857, Matthews correlation coefficient (MCC) = 0.6998). The method was trained and tested on a non redundant data set of 380 drugs (122 anticancer and 258 non-anticancer). The elaborated QSAR model based on the Artificial Neural Networks approach has been extensively validated and has confidently assigned anticancer character to a number of trial anticancer drugs from the literature.

Comparative QSAR- and fragments distribution analysis of drugs, druglikes, metabolic substances, and antimicrobial compounds

A number of binary QSAR models have been developed using methods of artificial neural networks, k-nearest neighbors, linear discriminative analysis, and multiple linear regression and have been compared for their ability to recognize five types of chemical compounds that include conventional drugs, inactive druglikes, antimicrobial substituents, and bacterial and human metabolites. Thus, 20 binary classifiers have been created using a variety of 'inductive' and traditional 2D QSAR descriptors which allowed up to 99% accurate separation of the studied groups of activities. The comparison of the performance by four computational approaches demonstrated that the neural nets result in generally more accurate predictions, followed closely by k-nearest neighbors methods. It has also been demonstrated that complementation of 'inductive' descriptors with conventional QSAR parameters does not generally improve the quality of resulting solutions, conforming high predictive ability of 'inductive' variables. The conducted comparative QSAR analysis based on a novel linear optimization approach has helped to identify the extent of overlapping between the studied groups of compounds, such as cross-recognition of bacterial metabolites and antimicrobial compounds reflecting their immanent resemblance and similar origin. Human metabolites have been characterized as a very distinctive class of substances, separated from all other groups in the descriptors space and exhibiting different QSAR behavior. The analysis of unique structural fragments and substituents revealed inhomogeneous scale-free organization of human metabolites illustrating the fact that certain molecular scaffolds (such as sugars and nucleotides) may be strongly favored by natural evolution. The established scale-free organization of human metabolites has been contemplated as a factor of their unique positioning in the descriptors space and their distinctive QSAR properties. It is anticipated that the study may bring additional insight into QSAR determinants for conventional drugs, inactive chemicals, and metabolic substances and may help in rationalizing design and discovery of novel antimicrobials and human therapeutics with improved, metabolite-like properties.

Design of Novispirin Antimicrobial Peptides by Quantitative Structure–Activity Relationship

Novispirin G10 is an alpha-helical antimicrobial peptide designed in an effort to develop alternative treatments against multidrug-resistant micro-organisms. To further optimize the antimicrobial activity, 58 novispirin analogs were constructed and used to establish a quantitative structure-activity relationship model. A statistically significant model (r2 = 0.73, q2 = 0.61) was obtained using a set of 69 selected molecular descriptors. Among these, VolSurf and charged partial surface area descriptors played a dominant role. Analysis of the model indicated that hydrophobicity, amphipathicity and charge were the most important features influencing activity for this set of peptides. Furthermore, the ability of the quantitative structure-activity relationship model to predict bioactivity was evaluated by analyzing a set of 400 novispirin analogs designed by molecular modeling. Out of these 400, 16 new novispirins with a higher predicted antimicrobial activity were tested in the suicide expression system, and about three out of four appeared more potent than the parent novispirin G10. Combination of VolSurf and charged partial surface area descriptors seems relevant to depict the interaction between novispirin and its target(s), presumably the microbial cell membrane. The presented findings show that modeling and quantitative structure-activity relationship methods can be useful in the construction of and/or optimization of the bioactivity of antimicrobial peptides for further development as effective antibiotic therapeutics.

Can ‘Bacterial-Metabolite-Likeness' Model Improve Odds of ‘in Silico' Antibiotic Discovery?

'Inductive' QSAR descriptors have been used to develop the series of QSAR models enabling 'in silico' distinguishing between antimicrobial compounds, conventional drugs, and druglike substances. The constructed neural network-based models operating by 30 'inductive' parameters have been validated on an extensive set of 2686 chemical structures and resulted in up to 97% accurate separation of the three types of molecular activities. The demonstrated ability of 'inductive' parameters to adequately capture molecular features determining 'antibiotic-like' and 'druglike' potentials have been further utilized to construct a model of 'Bacterial-Metabolite-Likeness' (BML). The same 'inductive' descriptors have been used to train a neural network that could very accurately recognize substances involved into bacterial metabolism (that have been experimentally identified). When the developed model has been applied to the mixed set of antimicrobials, drugs, and druglike chemicals (not used for training the BML model), it exhibited a 2-5-fold recognition preference toward antimicrobial compounds compared to general drugs and an 18- to 45-fold preference when compared to a druglike substance (depending on the model stringency). These results illustrate immanent similarity between conventional antimicrobials and native bacterial metabolites and suggest that the developed BML model can be an effective classification tool for 'in silico' antibiotic studies.

Successful in Silico Discovery of Novel Nonsteroidal Ligands for Human Sex Hormone Binding Globulin

Using "in silico" drug design methodologies, we have discovered several nonsteroidal compounds of natural origin that bind to human sex hormone binding globulin (SHBG) with affinity constants of 0.1 x 10(6) to 1.2 x 10(6) M(-1). The computational solutions we developed involved pharmacophore-aided database search, virtual protein-ligand docking, and structure-activity modeling with "inductive" QSAR descriptors. By screening 23 836 natural substance structures, we identified 29 potential SHBG ligands, and eight of these bound the protein in vitro. These nonsteroidal ligands belong to four classes of molecular scaffolds with several available substitution positions that could allow chemical modification to enhance SHBG-binding activity. Interestingly, one of these compounds is structurally similar to a dicyclohexane derivative that binds to rat SHBG and causes azospermia when administered to male rats. Taken together, the in silico strategy we have developed will aid in the discovery of nonsteroidal ligands of SHBG with novel pharmacological properties.