Prediction of IC50 values for ACAT inhibitors from molecular structure

Bio-activity of aminosulfonyl ureas in the light of nucleic acid bases and DNA base pair interaction

The quantum chemical descriptors based on density functional theory (DFT) are applied to predict the biological activity (log IC₅₀) of one class of acyl-CoA: cholesterol O-acyltransferase (ACAT) inhibitors, viz. aminosulfonyl ureas. ACAT are very effective agents for reduction of triglyceride and cholesterol levels in human body. Successful two parameter quantitative structure-activity relationship (QSAR) models are developed with a combination of relevant global and local DFT based descriptors for prediction of biological activity of aminosulfonyl ureas. The global descriptors, electron affinity of the ACAT inhibitors (EA) and/or charge transfer (ΔN) between inhibitors and model biosystems (NA bases and DNA base pairs) along with the local group atomic charge on sulfonyl moiety (∑Q_Sul) of the inhibitors reveals more than 90% efficacy of the selected descriptors for predicting the experimental log (IC₅₀) values.

AVP-IC50 Pred: Multiple machine learning techniques-based prediction of peptide antiviral activity in terms of half maximal inhibitory concentration (IC50)

Peptide-based antiviral therapeutics has gradually paved their way into mainstream drug discovery research. Experimental determination of peptides' antiviral activity as expressed by their IC50 values involves a lot of effort. Therefore, we have developed "AVP-IC50 Pred," a regression-based algorithm to predict the antiviral activity in terms of IC50 values (μM). A total of 759 non-redundant peptides from AVPdb and HIPdb were divided into a training/test set having 683 peptides (T(683)) and a validation set with 76 independent peptides (V(76)) for evaluation. We utilized important peptide sequence features like amino-acid compositions, binary profile of N8-C8 residues, physicochemical properties and their hybrids. Four different machine learning techniques (MLTs) namely Support vector machine, Random Forest, Instance-based classifier, and K-Star were employed. During 10-fold cross validation, we achieved maximum Pearson correlation coefficients (PCCs) of 0.66, 0.64, 0.56, 0.55, respectively, for the above MLTs using the best combination of feature sets. All the predictive models also performed well on the independent validation dataset and achieved maximum PCCs of 0.74, 0.68, 0.59, 0.57, respectively, on the best combination of feature sets. The AVP-IC50 Pred web server is anticipated to assist the researchers working on antiviral therapeutics by enabling them to computationally screen many compounds and focus experimental validation on the most promising set of peptides, thus reducing cost and time efforts. The server is available at http://crdd.osdd.net/servers/ic50avp.

Prediction of bioactivity of ACAT2 inhibitors by multilinear regression analysis and support vector machine

Two quantitative structure-activity relationships (QSAR) models for predicting 95 compounds inhibiting Acyl-coenzyme A: cholesterol acyltransferase2 (ACAT2) were developed. The whole data set was randomly split into a training set including 72 compounds and a test set including 23 compounds. The molecules were represented by 11 descriptors calculated by software ADRIANA.Code. Then the inhibitory activity of ACAT2 inhibitors was predicted using multilinear regression (MLR) analysis and support vector machine (SVM) method, respectively. The correlation coefficients of the models for the test sets were 0.90 for MLR model, and 0.91 for SVM model. Y-randomization was employed to ensure the robustness of the SVM model. The atom charge and electronegativity related descriptors were important for the interaction between the inhibitors and ACAT2.

Genetic algorithm optimization in drug design QSAR: Bayesian-regularized genetic neural networks (BRGNN) and genetic algorithm-optimized support vectors machines (GA-SVM)

Many articles in "in silico" drug design implemented genetic algorithm (GA) for feature selection, model optimization, conformational search, or docking studies. Some of these articles described GA applications to quantitative structure-activity relationships (QSAR) modeling in combination with regression and/or classification techniques. We reviewed the implementation of GA in drug design QSAR and specifically its performance in the optimization of robust mathematical models such as Bayesian-regularized artificial neural networks (BRANNs) and support vector machines (SVMs) on different drug design problems. Modeled data sets encompassed ADMET and solubility properties, cancer target inhibitors, acetylcholinesterase inhibitors, HIV-1 protease inhibitors, ion-channel and calcium entry blockers, and antiprotozoan compounds as well as protein classes, functional, and conformational stability data. The GA-optimized predictors were often more accurate and robust than previous published models on the same data sets and explained more than 65% of data variances in validation experiments. In addition, feature selection over large pools of molecular descriptors provided insights into the structural and atomic properties ruling ligand-target interactions.

Prediction of HPLC retention index using artificial neural networks and IGroup E-state indices

A back-propagation artificial neural network (ANN) was used to create a 10-fold leave-10%-out cross-validated ensemble model of high performance liquid chromatography retention index (HPLC-RI) for a data set of 498 diverse druglike compounds. A 10-fold multiple linear regression (MLR) ensemble model of the same data was developed for comparison. Molecular structure was described using IGroup E-state indices, a novel set of structure-information representation (SIR) descriptors, along with molecular connectivity chi and kappa indices and other SIR descriptors previously reported. The same input descriptors were used to develop models by both learning algorithms. The MLR model yielded marginally acceptable statistics with training correlation r(2) = 0.65, mean absolute error (MAE) = 83 RI units. External validation of 104 compounds not used for model development yielded validation v(2) = 0.49 and MAE = 73 RI units. The distribution of residuals for the fit and validate data sets suggest a nonlinear relationship between retention index and molecular structure as described by the SIR indices. Not surprisingly, the ANN model was significantly more accurate for both training and validation with training set r(2) = 0.93, MAE = 30 RI units and validation v(2) = 0.84, MAE = 41 RI units. For the ANN model, a total of 91% of validation predictions were within 100 RI units of the experimental value.

Artificial neural networks in biology and chemistry: the evolution of a new analytical tool

Once regarded as an eccentric and unpromising algorithm for the analysis of scientific data, the neural network has been developed in the last decade into a powerful computational tool. Its use now spans all areas of science, from the physical sciences and engineering to the life sciences and allied subjects. Applications range from the assessment of epidemiological data or the deconvolution of spectra to highly practical applications, such as the electronic nose. This introductory chapter considers briefly the growth in the use of neural networks and provides some general background in preparation for the more detailed chapters that follow.

Caco-2 cell permeability modelling: a neural network coupled genetic algorithm approach

The ability to cross the intestinal cell membrane is a fundamental prerequisite of a drug compound. However, the experimental measurement of such an important property is a costly and highly time consuming step of the drug development process because it is necessary to synthesize the compound first. Therefore, in silico modelling of intestinal absorption, which can be carried out at very early stages of drug design, is an appealing alternative procedure which is based mainly on multivariate statistical analysis such as partial least squares (PLS) and neural networks (NN). Our implementation of neural network models for the prediction of intestinal absorption is based on the correlation of Caco-2 cell apparent permeability (P (app)) values, as a measure of intestinal absorption, to the structures of two different data sets of drug candidates. Several molecular descriptors of the compounds were calculated and the optimal subsets were selected using a genetic algorithm; therefore, the method was indicated as Genetic Algorithm-Neural Network (GA-NN). A methodology combining a genetic algorithm search with neural network analysis applied to the modelling of Caco-2 P (app) has never been presented before, although the two procedures have been already employed separately. Moreover, we provide new Caco-2 cell permeability measurements for more than two hundred compounds. Interestingly, the selected descriptors show to possess physico-chemical connotations which are in excellent accordance with the well known relevant molecular properties involved in the cellular membrane permeation phenomenon: hydrophilicity, hydrogen bonding propensity, hydrophobicity and molecular size. The predictive ability of the models, although rather good for a preliminary study, is somewhat affected by the poor precision of the experimental Caco-2 measurements. Finally, the generalization ability of one model was checked on an external test set not derived from the data sets used to build the models. The result obtained is of interesting practical application and underlines that the successful model construction is strictly dependent on the structural space representation of the data set used for model development.

Novel approach to evolutionary neural network based descriptor selection and QSAR model development

Capability of evolutionary neural network (ENN) based QSAR approach to direct the descriptor selection process towards stable descriptor subset (DS) composition characterized by acceptable generalization, as well as the influence of description stability on QSAR model interpretation have been examined. In order to analyze the DS stability and QSAR model generalization properties multiple random dataset partitions into training and test set were made. Acceptability criteria proposed by Golbraikh et al. [J. Comput.-Aided Mol. Des., 17 (2003) 241] have been chosen for selection of highly predictive QSAR models from a set of all models produced by ENN for each dataset splitting. All QSAR models that pass Golbraikh's filter generated by ENN for each dataset partition were collected. Two final DS forming principles were compared. Standard principle is based on selection of descriptors characterized by highest frequencies among all descriptors that appear in the pool [J. Chem. Inf. Comput. Sci., 43 (2003) 949]. Search across the model pool for DS that are stable against multiple dataset subsampling i.e. universal DS solutions is the basis of novel approach. Based on described principles benzodiazepine QSAR has been proposed and evaluated against results reported by others in terms of final DS composition and model predictive performance.

A unique lysophospholipid acyltransferase (LPAT) antagonist, CI-976, affects secretory and endocytic membrane trafficking pathways

Previous studies have shown that inhibition of a Golgi-complex-associated lysophospholipid acyltransferase (LPAT) activity by the drug CI-976 stimulates Golgi tubule formation and subsequent redistribution of resident Golgi proteins to the endoplasmic reticulum (ER). Here, we show that CI-976 stimulates tubule formation from all subcompartments of the Golgi complex, and often these tubules formed independently, i.e. individual tubules usually did not contain markers from different subcompartments. Whereas the cis, medial and trans Golgi membranes redistributed to the ER, the trans Golgi network (TGN) collapsed back to a compact juxtanuclear position similar to that seen with brefeldin A (BFA) treatment. Also similar to BFA, CI-976 induced the formation of endosome tubules, but unlike BFA, these tubules did not fuse with TGN tubules. Finally, CI-976 produced an apparently irreversible block in the endocytic recycling pathway of transferrin (Tf) and Tf receptors (TfRs) but had no direct effect on Tf uptake from the cell surface. Tf and TfRs accumulated in centrally located, Rab11-positive vesicles indicating that CI-976 inhibits export of cargo from the central endocytic recycling compartment. These results, together with previous studies, demonstrate that CI-976 inhibits multiple membrane trafficking steps, including ones found in the endocytic and secretory pathways, and imply a wider role for lysophospholipid acyltransferases in membrane trafficking.

Use of Acyltransferase Inhibitors to Block Vesicular Traffic Between the ER and Golgi Complex

This article describes the use of acyltransferase inhibitors as probes for studying the potential role of lysophospholipid acyltransferases (LPAT) in intracellular membrane trafficking in the secretory and endocytic pathways. The small molecule inhibitors that are described here were originally found as acyl-CoA:cholesterol acyltransferase (ACAT) inhibitors. One of these, CI-976 (2,2-methyl-N-(2,4,6,-trimethoxyphenyl)dodecanamide), was also found to be a potent LPAT inhibitor. CI-976 is a small, hydrophobic, membrane-permeant compound and both in vivo and in vitro studies have shown that it, but not other ACAT inhibitors, has a profound effect on multiple membrane trafficking pathways in eukaryotic cells including: (1) inhibition of COPII vesicle budding from the endoplasmic reticulum (ER), (2) inhibition of transferrin and transferrin receptor export from the endocytic recycling compartment, and (3) stimulation of tubule-mediated retrograde trafficking of Golgi membranes to the ER. Here we describe the use of CI-976 and other ACAT inhibitors for studies with both cultured mammalian cells and in vitro reconstitution assays, with a particular emphasis on COPII vesicle budding from the ER. All of these studies strongly suggest that CI-976-sensitive LPATs play a role in coated vesicle fission, and therefore, CI-976 is a valuable addition to the arsenal of small molecule inhibitors that can be used to study secretory and endocytic membrane trafficking pathways.

Development of QSAR Models To Predict and Interpret the Biological Activity of Artemisinin Analogues

This work presents the development of Quantitative Structure-Activity Relationship (QSAR) models to predict the biological activity of 179 artemisinin analogues. The structures of the molecules are represented by chemical descriptors that encode topological, geometric, and electronic structure features. Both linear (multiple linear regression) and nonlinear (computational neural network) models are developed to link the structures to their reported biological activity. The best linear model was subjected to a PLS analysis to provide model interpretability. While the best linear model does not perform as well as the nonlinear model in terms of predictive ability, the application of PLS analysis allows for a sound physical interpretation of the structure-activity trend captured by the model. On the other hand, the best nonlinear model is superior in terms of pure predictive ability, having a training error of 0.47 log RA units (R2 = 0.96) and a prediction error of 0.76 log RA units (R2 = 0.88).

Inhibition of a Golgi complex lysophospholipid acyltransferase induces membrane tubule formation and retrograde trafficking

Recent studies have suggested that formation of Golgi membrane tubules involves the generation of membrane-associated lysophospholipids by a cytoplasmic Ca2+-independent phospholipase A2 (PLA2). Herein, we provide additional support for this idea by showing that inhibition of lysophospholipid reacylation by a novel Golgi-associated lysophosphatidylcholine acyltransferase (LPAT) induces the rapid tubulation of Golgi membranes, leading in their retrograde movement to the endoplasmic reticulum. Inhibition of the Golgi LPAT was achieved by 2,2-dimethyl-N-(2,4,6-trimethoxyphenyl)dodecanamide (CI-976), a previously characterized antagonist of acyl-CoA cholesterol acyltransferase. The effect of CI-976 was similar to that of brefeldin A, except that the coatomer subunit beta-COP remained on Golgi-derived membrane tubules. CI-976 also enhanced the cytosol-dependent formation of tubules from Golgi complexes in vitro and increased the levels of lysophosphatidylcholine in Golgi membranes. Moreover, preincubation of cells with PLA2 antagonists inhibited the ability of CI-976 to induce tubules. These results suggest that Golgi membrane tubule formation can result from increasing the content of lysophospholipids in membranes, either by stimulation of a PLA2 or by inhibition of an LPAT. These two opposing enzyme activities may help to coordinately regulate Golgi membrane shape and tubule formation.

Classification of inhibitors of protein tyrosine phosphatase 1B using molecular structure based descriptors

Loss of Protein Tyrosine Phosphatase 1B (PTP 1B) activity is known to enhance insulin sensitivity and resistance to weight gain. So potent and orally active PTP1B inhibitors could be potential pharmacological agents for the treatment of Type 2 diabetes and obesity. Classification models of PTP1B inhibitors are developed using a data set containing 128 compounds. Their inhibitory concentrations ranged from -1.59 to 1.68 log units. Initially a two-class (active, inactive) problem is tackled using a number of different methods. The data set was divided into active and inactive classes on the basis of inhibitory activity of the compounds. Molecular structure-based descriptors were calculated and used in the model development. Descriptors encoding the flexibility of the molecules were investigated. Classification models were generated using k-nearest neighbors (k-NN), linear discriminant analysis (LDA), and radial basis function neural network (RBFNN). All models are tested using an external prediction set, compounds not used anywhere during the model development procedure. A five-descriptor model is developed that produces a classification rate of 85.7% for an external prediction set. Then a three-class (active, moderately active, inactive) problem was explored. This time the data set was divided into highly active, moderate, and inactive classes on the basis of inhibitory activity of the compounds. The best classification rate achieved for an external prediction set was 85%. The classification rates achieved indicate that these models could serve as a screening mechanism, to identify potentially useful PTP 1B inhibitors. In addition multiple linear regression and computational neural network models are also developed for prediction of log IC(50) values. All QSAR models are tested using the same external prediction set.

E-state modeling of fish toxicity independent of 3D structure information

Topological structure methods are used to model fish toxicity against three classes of organic chemicals. The models were obtained independent of 3D structure information. Further, no mechanism of partitioning was assumed, thus avoiding the problems associated with selection of partitioning system for computation of log P. QSAR models were developed for a set of 92 compounds, including phenols, anilines and substituted aromatic hydrocarbons, yielding excellent statistics: r2 = 0.87, s = 0.25 and q2 = 0.85 leave-one-out (LOO), that are better than those reported in the literature. The model is based on molecular connectivity valence chi-1 index [1chiv], the atom type E-State indices for chlorine [ST(-Cl)] and for ether oxygen [ST(-O-)], and the maximum hydrogen E-State atom value in a molecule [Hmax]. Each of the subgroups was also separately well modeled. The model for the full set is validated through use of external validation test sets and ten-fold cross-validation (repeated three times). The quality of the validation statistics supports the claim that the model may be used for estimation of pLC50 values for similar molecules. Detailed structure interpretation is given for the descriptors in the model. These four structure descriptors encode influence of molecular context of groups as well as counts of those groups, in addition to molecular skeletal structure.

Prediction of dihydrofolate reductase inhibition and selectivity using computational neural networks and linear discriminant analysis

A data set of 345 dihydrofolate reductase inhibitors was used to build QSAR models that correlate chemical structure and inhibition potency for three types of dihydrofolate reductase (DHFR): rat liver (rl), Pneumocystis carinii (pc), and Toxoplasma gondii (tg). Quantitative models were built using subsets of molecular structure descriptors being analyzed by computational neural networks. Neural network models were able to accurately predict log IC(50) values for the three types of DHFR to within +/-0.65 log units (data sets ranged approximately 5.5 log units) of the experimentally determined values. Classification models were also constructed using linear discriminant analysis to identify compounds as selective or nonselective inhibitors of bacterial DHFR (pcDHFR and tgDHFR) relative to mammalian DHFR (rlDHFR). A leave-N-out training procedure was used to add robustness to the models and to prove that consistent results could be obtained using different training and prediction set splits. The best linear discriminant analysis (LDA) models were able to correctly predict DHFR selectivity for approximately 70% of the external prediction set compounds. A set of new nitrogen and oxygen-specific descriptors were developed especially for this data set to better encode structural features, which are believed to directly influence DHFR inhibition and selectivity.

Prediction of glycine/NMDA receptor antagonist inhibition from molecular structure

The design and blood brain barrier crossing of glycine/NMDA receptor antagonists are of significant interest in pharmaceutical research. The use of these antagonists in stroke or seizure reduction have been considered. Measuring the inhibitory concentrations, however, can be time-consuming and costly. The use of quantitative structure-activity relationships to estimate IC(50) values for these receptor antagonists is an attractive alternative compared to experimental measurement. A data set of 109 compounds with measured log(IC(50)) values ranging from -0.57 to 4.5 is used. Structural information is encoded with numerical descriptors for topological, electronic, geometric, and polar surface properties. A genetic algorithm with a computational neural network fitness evaluator is used to select the best descriptor subsets. Multiple linear regression and computational neural network models are developed. Additionally, a quantitative radial basis function neural network (QRBFNN) was developed with the intent of introducing nonlinearity at a faster speed. A genetic algorithm using the radial basis function network as a fitness evaluator was also developed to search descriptor space for optimum subsets. All models are tested using an external prediction set. The nonlinear computational neural network model has root-mean-square errors of approximately half a log unit.

E-state modeling of HIV-1 protease inhibitor binding independent of 3D information

Data for HIV-1 protease inhibitors (in vitro enzyme binding) were used as a training set to develop a QSAR model based on topological descriptors, including two hydrogen E-state indices, along with a molecular connectivity chi and a kappa shape index. A statistically satisfactory four-variable model was obtained for the 32 compounds in the training set, r2 = 0.86, s = 0.60, and q2 = 0.79, without the use of information from 3D geometries or detailed interaction energy calculations. The model was validated through the prediction of 15 compounds in the external test set, yielding a mean absolute error, MAE, = 0.82. Structure interpretation is given for each variable to assist in the design of new compounds. Structure features emphasized in the model include hydrogen bond donating ability, nonpolar groups, skeletal branching, and molecular globularity. On the basis of these statistical criteria, this E-state model may be considered useful for prediction of pIC50 values for new HIV-1 protease inhibitors.

Development of quantitative structure-activity relationship and classification models for a set of carbonic anhydrase inhibitors

Mathematical models are developed to find quantitative structure-activity relationships that correlate chemical structure and inhibition toward three carbonic anhydrase (CA) isozymes: CA I, II, and IV. Numerical descriptors are generated to encode important topological, geometric, and electronic features of molecular structure. After descriptor generation, multiple linear regression, and computational neural network (CNN) analyses are performed on various descriptor subsets to find superior models for prediction. Committees of five CNNs were utilized to average final predicted values for the 142-compound data set. For inhibitors of CA I, an 8-5-1 CNN committee produced a training set rms error of 0.105 log K(i) (r(2) = 0.994) and prediction set rms error of 0.208 log K(i) (r(2) = 0.980). Training and prediction set rms errors of 0.140 log K(i) (r(2) = 0.992) and 0.231 log K(i) (r(2) = 0.971), respectively, were produced by a 9-5-1 CNN committee for inhibitors of CA II. For prediction of CA IV inhibitors, an 8-5-1 CNN committee produced training and prediction set rms errors of 0.147 log K(i) (r(2) = 0.992) and 0.211 log K(i) (r(2) = 0.991), respectively. In addition, classification models were built using k-nearest neighbor (kNN) analysis to solve two- and three-class problems for inhibitors of CA IV. A three-descriptor classification model proved superior in labeling compounds as active or inactive inhibitors for the two-class problem. Training and prediction set percent classification rates of 100% and 87.1%, respectively, were obtained. For the three-class (active/moderate/inactive) problem, a five-descriptor model was deemed optimal producing a training set percent classification rate of 98.8% and prediction set rate of 79.0%.

Linear regression and computational neural network prediction of tetrahymena acute toxicity for aromatic compounds from molecular structure

A quantitative structure toxicity relationship (QSTR) has been derived for a diverse set of 448 industrially important aromatic solvents. Toxicity was expressed as the 50% growth impairment concentration (ICG(50)) for the ciliated protozoa Tetrahymena and spans the range -1.46 to 3.36 log units. Molecular descriptors that encode topological, geometrical, electronic, and hybrid geometrical-electronic structural features were calculated for each compound. Subsets of molecular descriptors were selected via a simulated annealing technique and a genetic algorithm. From this reduced pool of descriptors, multiple linear regression models and nonlinear models using computational neural networks (CNNs) were derived and then used to predict the ICG(50) values for an external set of representative compounds. An average of 10 nonlinear CNN models with 11-5-1 architecture was found to best describe the system with root-mean-square errors of 0.28, 0.29, and 0.34 log units for the training, cross validation, and prediction sets, respectively.

The impact of informatics and computational chemistry on synthesis and screening

High-throughput synthesis and screening technologies have enhanced the impact of computational chemistry on the drug discovery process. From the design of targeted, drug-like libraries to 'virtual' optimization of potency, selectivity and ADME/Tox properties, computational chemists are able to efficiently manage costly resources and dramatically shorten drug discovery cycle times. This review will describe some of the successful strategies and applications of state-of-the-art algorithms to enhance drug discovery, as well as key points in the drug discovery process where computational methods can have, and have had, greatest impact.

QSAR and k-nearest neighbor classification analysis of selective cyclooxygenase-2 inhibitors using topologically-based numerical descriptors

Experimental IC(50) data for 314 selective cyclooxygenase-2 (COX-2) inhibitors are used to develop quantitation and classification models as a potential screening mechanism for larger libraries of target compounds. Experimental log(IC(50)) values ranged from 0.23 to > or = 5.00. Numerical descriptors encoding solely topological information are calculated for all structures and are used as inputs for linear regression, computational neural network, and classification analysis routines. Evolutionary optimization algorithms are then used to search the descriptor space for information-rich subsets which minimize the rms error of a diverse training set of compounds. An eight-descriptor model was identified as a robust predictor of experimental log(IC(50)) values, producing a root-mean-square error of 0.625 log units for an external prediction set of inhibitors which took no part in model development. A k-nearest neighbor classification study of the data set discriminating between active and inactive members produced a nine-descriptor model able to accurately classify 83.3% of the prediction set compounds correctly.

Prediction of aqueous solubility of heteroatom-containing organic compounds from molecular structure

The use of quantitative structure-property relationships (QSPRs) to predict aqueous solubilities (log S) of heteroatom-containing organic compounds from their molecular structure is presented. Three data sets are examined. Data set 1 contains 176 compounds having one or more nitrogen atoms with some oxygen (log S[mol/L] range is -7.41 to 0.96). Data set 2 contains 223 compounds having one or more oxygen atoms, with no nitrogen (log S[mol/L] range is -8.77 to 1.57). Data set 3 contains all 399 compounds from sets 1 and 2 (log S/mol/L] range is -8.77 to 1.57). After descriptor generation and feature selection, multiple linear regression (MLR) and computational neural network (CNN) models are developed for aqueous solubility prediction. The best results were obtained with nonlinear CNN models. Root-mean-square (rms) errors for training with the three data sets ranged from 0.3 to 0.6 log units. All models were validated with external prediction sets, with the rms errors ranging from 0.6 log units to 1.5 log units.