Corchop – an Interactive Routine for the Dimension Reduction of Large QSAR Data Sets

Study of the antimalarial activity of 4-aminoquinoline compounds against chloroquine-sensitive and chloroquine-resistant parasite strains

This study is concerned with identifying features of 4-aminoquinoline scaffolds that can help pinpoint characteristics that enhance activity against chloroquine-resistant parasites. Statistically valid predictive models are reported for a series of 4-aminoquinoline analogues that are active against chloroquine-sensitive (NF54) and chloroquine-resistant (K1) strains of Plasmodium falciparum. Quantitative structure activity relationship techniques, based on statistical and machine learning methods such as multiple linear regression and partial least squares, were used with a novel pruning method for the selection of descriptors to develop robust models for both strains. Inspection of the dominant descriptors supports the hypothesis that chemical features that enable accumulation in the food vacuole of the parasite are key determinants of activity against both strains. The hydrophilic properties of the compounds were found to be crucial in predicting activity against the chloroquine-sensitive NF54 parasite strain, but not in the case of the chloroquine-resistant K1 strain, in line with previous studies. Additionally, the models suggest that ‘softer’ compounds tend to have improved activity for both strains than do ‘harder’ ones. The internally and externally validated models reported here should also prove useful in the future screening of potential antimalarial compounds for targeting chloroquine-resistant strains.

Graphical Abstract

Second-generation nitazoxanide derivatives: thiazolides are effective inhibitors of the influenza A virus

AIM

The only small molecule drugs currently available for treatment of influenza A virus (IAV) are M2 ion channel blockers and sialidase inhibitors. The prototype thiazolide, nitazoxanide, has successfully completed Phase III clinical trials against acute uncomplicated influenza.

RESULTS

We report the activity of seventeen thiazolide analogs against A/PuertoRico/8/1934(H1N1), a laboratory-adapted strain of the H1N1 subtype of IAV, in a cell culture-based assay. A total of eight analogs showed IC₅₀s in the range of 0.14-5.0 μM. Additionally a quantitative structure-property relationship study showed high correlation between experimental and predicted activity based on a molecular descriptor set.

CONCLUSION

A range of thiazolides show useful activity against an H1N1 strain of IAV. Further evaluation of these molecules as potential new small molecule therapies is justified.

In silico modelling of permeation enhancement potency in Caco-2 monolayers based on molecular descriptors and random forest

Structural traits of permeation enhancers are important determinants of their capacity to promote enhanced drug absorption. Therefore, in order to obtain a better understanding of structure-activity relationships for permeation enhancers, a Quantitative Structural Activity Relationship (QSAR) model has been developed. The random forest-QSAR model was based upon Caco-2 data for 41 surfactant-like permeation enhancers from Whitehead et al. (2008) and molecular descriptors calculated from their structure. The QSAR model was validated by two test-sets: (i) an eleven compound experimental set with Caco-2 data and (ii) nine compounds with Caco-2 data from literature. Feature contributions, a recent developed diagnostic tool, was applied to elucidate the contribution of individual molecular descriptors to the predicted potency. Feature contributions provided easy interpretable suggestions of important structural properties for potent permeation enhancers such as segregation of hydrophilic and lipophilic domains. Focusing on surfactant-like properties, it is possible to model the potency of the complex pharmaceutical excipients, permeation enhancers. For the first time, a QSAR model has been developed for permeation enhancement. The model is a valuable in silico approach for both screening of new permeation enhancers and physicochemical optimisation of surfactant enhancer systems.

Thiazolides as novel antiviral agents. 2. Inhibition of hepatitis C virus replication

We report the activities of a number of thiazolides [2-hydroxyaroyl-N-(thiazol-2-yl)amides] against hepatitis C virus (HCV) genotypes IA and IB, using replicon assays. The structure-activity relationships (SARs) of thiazolides against HCV are less predictable than against hepatitis B virus (HBV), though an electron-withdrawing group at C(5') generally correlates with potency. Among the related salicyloylanilides, the m-fluorophenyl analogue was most promising; niclosamide and close analogues suffered from very low solubility and bioavailability. Nitazoxanide (NTZ) 1 has performed well in clinical trials against HCV. We show here that the 5'-Cl analogue 4 has closely comparable in vitro activity and a good cell safety index. By use of support vector analysis, a quantitative structure-activity relationship (QSAR) model was obtained, showing good predictive models for cell safety. We conclude by updating the mode of action of the thiazolides and explain the candidate selection that has led to compound 4 entering preclinical development.

Thiazolides as novel antiviral agents. 1. Inhibition of hepatitis B virus replication

We report the syntheses and activities of a wide range of thiazolides [viz., 2-hydroxyaroyl-N-(thiazol-2-yl)amides] against hepatitis B virus replication, with QSAR analysis of our results. The prototypical thiazolide, nitazoxanide [2-hydroxybenzoyl-N-(5-nitrothiazol-2-yl)amide, NTZ] 1 is a broad spectrum antiinfective agent effective against anaerobic bacteria, viruses, and parasites. By contrast, 2-hydroxybenzoyl-N-(5-chlorothiazol-2-yl)amide 3 is a novel, potent, and selective inhibitor of hepatitis B replication (EC(50) = 0.33 μm) but is inactive against anaerobes. Several 4'- and 5'-substituted thiazolides show good activity against HBV; by contrast, some related salicyloylanilides show a narrower spectrum of activity. The ADME properties of 3 are similar to 1; viz., the O-acetate is an effective prodrug, and the O-aryl glucuronide is a major metabolite. The QSAR study shows a good correlation of observed EC(90) for intracellular virions with thiazolide structural parameters. Finally we discuss the mechanism of action of thiazolides in relation to the present results.

Adaptive neuro-fuzzy inference system (ANFIS): A new approach to predictive modeling in QSAR applications: A study of neuro-fuzzy modeling of PCP-based NMDA receptor antagonists

This paper proposes a new method, Adaptive Neuro-Fuzzy Inference System (ANFIS) to evaluate physicochemical descriptors of certain chemical compounds for their appropriate biological activities in terms of QSAR models with the aid of artificial neural network (ANN) approach combined with the principle of fuzzy logic. The ANFIS was utilized to predict NMDA (N-methyl-d-Aspartate) receptor binding activities of phencyclidine (PCP) derivatives. A data set of 38 drug-like compounds was coded with 1244 calculated molecular structure descriptors (clustered in 20 data sets) which were obtained from several sources, mainly from Dragon software. Prior to the progress to the ANFIS system, descriptors from the best subsets were selected using unsupervised forward selection (UFS) to eliminate redundancy and multicollinearity followed by fuzzy linear regression algorithm (FLR) which was used for variable selection. ANFIS was applied to train the final descriptors (Mor22m, E3s, R3v+, and R1e+) using a hybrid algorithm consisting of back-propagation and least-square estimation while the optimum number and shape of related functions were obtained through the subtractive clustering algorithm. Comparison of the proposed method with traditional methods, that is, multiple linear regression (MLR) and partial least-square (PLS) was also studied and the results indicated that the ANFIS model obtained from data sets achieved satisfactory accuracy.

Comparison of commercially available genetic algorithms: GAs as variable selection tool

Many commercially available software programs claim similar efficiency and accuracy as variable selection tools. Genetic algorithms are commonly used variable selection methods where most relevant variables can be differentiated from 'less important' variables using evolutionary computing techniques. However, different vendors offer several algorithms, and the puzzling question is: which one is the appropriate method of choice? In this study, several genetic algorithm tools (e.g. GFA from Cerius2, QuaSAR-Evolution from MOE and Partek's genetic algorithm) were compared. Stepwise multiple linear regression models were generated using the most relevant variables identified by the above genetic algorithms. This procedure led to the successful generation of Quantitative Structure activity Relationship (QSAR) models for (a) proprietary datasets and (b) the Selwood dataset.

A method for quantifying and visualizing the diversity of QSAR models

Feature selection is one of the most commonly used and reliable methods for deriving predictive quantitative structure-activity relationships (QSAR). Many feature selection algorithms are stochastic in nature and often produce different solutions depending on the initialization conditions. Because some features may be highly correlated, models that are based on different sets of descriptors may capture essentially the same information, however, such models are difficult to recognize. Here, we introduce a measure of similarity between QSAR models that captures the correlation between the underlying features. This measure can be used in conjunction with stochastic proximity embedding (SPE) or multi-dimensional scaling (MDS) to create a meaningful visual representation of structure-activity model space and aid in the post-processing and analysis of results of feature selection calculations.

Variable selection using pair-correlation method. Environmental applications

Pair-correlation method (PCM) has been developed for selecting between two, correlated descriptor variables. PCM utilizes systematic information present in the scatter of QSAR applications. The data are suitably ordered in a 2 x 2 contingency table. Statistical tests are used to discriminate between the descriptor variables. We have developed, adapted, investigated and compared the following test statistics to each other: Conditional Fisher's exact test (CE), McNemar's test (MN), Chi-square test and Williams' t-test (Wt). If a test indicates significant difference between the descriptors, we use the terms superior-inferior or winner-loser for the overwhelming and subordinate descriptors, respectively. If more than two variables are to be compared, the discrimination can be made pair-wise and then the variables have to be ordered. Three ways of ordering have been used: simple ordering (number of wins), ordering according to the differences between wins and losses, and ordering according to probability-weighted differences between wins and losses. The basic algorithm of PCM has been described in this paper, the various selection criteria and the ordering methods were compared on suitable model systems. These case studies involved description of cAMP phospodiesterase inhibition by flavons, toxicity of chlorobenzenes and mutagenic character of aromatic and heteroaromatic amines.

Modelling mutagenicity using properties calculated by computational chemistry

The recent advances in combinatorial chemistry and high throughput screening technologies have led to an explosion in the numbers of possible therapeutic candidates being produced at the early stages of drug discovery. This rapid increase in the number of chemicals to be classified results in a greater need for alternative methods for the prediction of toxicity. Most QSAR models for mutagenicity have been constructed for congeneric series. The prediction requirements of the pharmaceutical industry, however, cover quite diverse chemical structures. This paper reports a study of mutagenicity data for a diverse set of 90 compounds. Good discriminant models have been built for this data set using properties calculated by the techniques of computational chemistry. Jack-knifed (leave one out) predictions for these models are of the order of 85%.

Unsupervised forward selection: a method for eliminating redundant variables

An unsupervised learning method is proposed for variable selection and its performance assessed using three typical QSAR data sets. The aims of this procedure are to generate a subset of descriptors from any given data set in which the resultant variables are relevant, redundancy is eliminated, and multicollinearity is reduced. Continuum regression, an algorithm encompassing ordinary least squares regression, regression on principal components, and partial least squares regression, was used to construct models from the selected variables. The variable selection routine is shown to produce simple, robust, and easily interpreted models for the chosen data sets.

Pattern recognition methods in rational drug design

Pattern recognition methods have much to offer the drug designer, particularly as the calculation and collation of data, both biological and physicochemical, becomes easier with the widespread use of computer databases, molecular modeling systems, and property prediction packages. Some of the techniques, however, suffer from difficulties in interpretation and the dangers of chance effects have received little attention. The wider use and understanding of these methods is expected to enhance their utility in drug design. Finally, it should be mentioned here that these methods are becoming applied increasingly in other areas of pharmaceutical research, e.g., the analysis of clinical data, and that new techniques for analysis continue to be developed and applied in this field.