Ze-Rong Li

Update of PROFEAT: a web server for computing structural and physicochemical features of proteins and peptides from amino acid sequence

Sequence-derived structural and physicochemical features have been extensively used for analyzing and predicting structural, functional, expression and interaction profiles of proteins and peptides. PROFEAT has been developed as a web server for computing commonly used features of proteins and peptides from amino acid sequence. To facilitate more extensive studies of protein and peptides, numerous improvements and updates have been made to PROFEAT. We added new functions for computing descriptors of protein–protein and protein–small molecule interactions, segment descriptors for local properties of protein sequences, topological descriptors for peptide sequences and small molecule structures. We also added new feature groups for proteins and peptides (pseudo-amino acid composition, amphiphilic pseudo-amino acid composition, total amino acid properties and atomic-level topological descriptors) as well as for small molecules (atomic-level topological descriptors). Overall, PROFEAT computes 11 feature groups of descriptors for proteins and peptides, and a feature group of more than 400 descriptors for small molecules plus the derived features for protein–protein and protein–small molecule interactions. Our computational algorithms have been extensively tested and used in a number of published works for predicting proteins of specific structural or functional classes, protein–protein interactions, peptides of specific functions and quantitative structure activity relationships of small molecules. PROFEAT is accessible free of charge at http://bidd.cz3.nus.edu.sg/cgi-bin/prof/protein/profnew.cgi.

Effect of Molecular Descriptor Feature Selection in Support Vector Machine Classification of Pharmacokinetic and Toxicological Properties of Chemical Agents

Statistical-learning methods have been developed for facilitating the prediction of pharmacokinetic and toxicological properties of chemical agents. These methods employ a variety of molecular descriptors to characterize structural and physicochemical properties of molecules. Some of these descriptors are specifically designed for the study of a particular type of properties or agents, and their use for other properties or agents might generate noise and affect the prediction accuracy of a statistical learning system. This work examines to what extent the reduction of this noise can improve the prediction accuracy of a statistical learning system. A feature selection method, recursive feature elimination (RFE), is used to automatically select molecular descriptors for support vector machines (SVM) prediction of P-glycoprotein substrates (P-gp), human intestinal absorption of molecules (HIA), and agents that cause torsades de pointes (TdP), a rare but serious side effect. RFE significantly reduces the number of descriptors for each of these properties thereby increasing the computational speed for their classification. The SVM prediction accuracies of P-gp and HIA are substantially increased and that of TdP remains unchanged by RFE. These prediction accuracies are comparable to those of earlier studies derived from a selective set of descriptors. Our study suggests that molecular feature selection is useful for improving the speed and, in some cases, the accuracy of statistical learning methods for the prediction of pharmacokinetic and toxicological properties of chemical agents.

PEARLS: program for energetic analysis of receptor-ligand system.

Analysis of the energetics of small molecule ligand-protein, ligand-nucleic acid, and protein-nucleic acid interactions facilitates the quantitative understanding of molecular interactions that regulate the function and conformation of proteins. It has also been extensively used for ranking potential new ligands in virtual drug screening. We developed a Web-based software, PEARLS (Program for Energetic Analysis of Ligand-Receptor Systems), for computing interaction energies of ligand-protein, ligand-nucleic acid, protein-nucleic acid, and ligand-protein-nucleic acid complexes from their 3D structures. AMBER molecular force field, Morse potential, and empirical energy functions are used to compute the van der Waals, electrostatic, hydrogen bond, metal-ligand bonding, and water-mediated hydrogen bond energies between the binding molecules. The change in the solvation free energy of molecular binding is estimated by using an empirical solvation free energy model. Contribution from ligand conformational entropy change is also estimated by a simple model. The computed free energy for a number of PDB ligand-receptor complexes were studied and compared to experimental binding affinity. A substantial degree of correlation between the computed free energy and experimental binding affinity was found, which suggests that PEARLS may be useful in facilitating energetic analysis of ligand-protein, ligand-nucleic acid, and protein-nucleic acid interactions. PEARLS can be accessed at http://ang.cz3.nus.edu.sg/cgi-bin/prog/rune.pl.

Efficacy of different protein descriptors in predicting protein functional families

BackgroundSequence-derived structural and physicochemical descriptors have frequently been used in machine learning prediction of protein functional families, thus there is a need to comparatively evaluate the effectiveness of these descriptor-sets by using the same method and parameter optimization algorithm, and to examine whether the combined use of these descriptor-sets help to improve predictive performance. Six individual descriptor-sets and four combination-sets were evaluated in support vector machines (SVM) prediction of six protein functional families.ResultsThe performance of these descriptor-sets were ranked by Matthews correlation coefficient (MCC), and categorized into two groups based on their performance. While there is no overwhelmingly favourable choice of descriptor-sets, certain trends were found. The combination-sets tend to give slightly but consistently higher MCC values and thus overall best performance such that three out of four combination-sets show slightly better performance compared to one out of six individual descriptor-sets.ConclusionOur study suggests that currently used descriptor-sets are generally useful for classifying proteins and the prediction performance may be enhanced by exploring combinations of descriptors.

Kinetics and Mechanism for Formation of Enols in Reaction of Hydroxide Radical with Propene

Recently, enols have been found to be the common intermediates in hydrocarbon combustion flames (Taatjes et al. Science 2005, 308, 1887), but the knowledge of kinetic properties for such species in combustion flames is rare. Therefore in this work, particular attention is paid to the formation of enols in combustion flames. Starting with HO and propene (CH(3)CH=CH(2)), the reaction mechanism involving eight product channels has been investigated systematically. It is revealed that the electrophilic addition of OH to the double bond of CH(3)CH=CH(2) is unselective and the chemically activated adducts, CH(3)CHOH=CH(2) and CH(3)CH=CH(2)OH, may undergo dissociation in competition with H-abstractions. The kinetics and product branching ratios of the HO and propene reaction have been evaluated in the temperature range of 200-3000 K by Variflex code, based on the weak collision master equation/microcanonical variational RRKM theory. Available experimental kinetic data can be quantitatively reproduced by this study, with a minor adjustment (1.0 kcal/mol) of the OH central addition barrier. From the theoretical calculations with multiple reflection correction included, the total rate constant is fitted to k(t) = 6.07 x 10(-5)T(-2.54) exp(108/T) cm(3) x molecule(-1) x s(-1) in the range of 200-800 K and k(t) = 7.11 x 10(-23)T(3.38) exp(-1097/T) cm(3) x molecule(-1) x s(-1) in the range of 800-3000 K, which are in close agreement with experimental data. The branching ratios of enol channels are consistent with the observation in low-pressure flames and hence the reaction mechanisms presented here provide valuable descriptions of enol formations in hydrocarbon combustion chemistry.

Prediction of chemical carcinogenicity by machine learning approaches

In this paper we report a successful application of machine learning approaches to the prediction of chemical carcinogenicity. Two different approaches, namely a support vector machine (SVM) and artificial neural network (ANN), were evaluated for predicting chemical carcinogenicity from molecular structure descriptors. A diverse set of 844 compounds, including 600 carcinogenic (CG+) and 244 noncarcinogenic (CG−) molecules, was used to estimate the accuracies of these approaches. The database was divided into two sets: the model construction set and the independent test set. Relevant molecular descriptors were selected by a hybrid feature selection method combining Fischers score and Monte Carlo simulated annealing from a wide set of molecular descriptors, including physiochemical properties, constitutional, topological, and geometrical descriptors. The first model validation method was based a five-fold cross-validation method, in which the model construction set is split into five subsets. The five-fold cross-validation was used to select descriptors and optimise the model parameters by maximising the averaged overall accuracy. The final SVM model gave an averaged prediction accuracy of 90.7% for CG+ compounds, 81.6% for CG− compounds and 88.1% for the overall accuracy, while the corresponding ANN model provided an averaged prediction accuracy of 86.1% for CG+ compounds, 79.3% for CG− compounds and 84.2% for the overall accuracy. These results indicate that the hybrid feature selection method is very efficient and the selected descriptors are truly relevant to the carcinogenicity of compounds. Another model validation method, i.e. a hold-out method, was used to build the classification model using the selected descriptors and the optimised model parameters, in which the whole model construction set was used to build the classification model and the independent test set was used to test the predictive ability of the model. The SVM model gave a prediction accuracy of 87.6% for CG+ compounds, 79.1% for CG− compounds and 85.0% for the overall accuracy. The ANN model gave a prediction accuracy of 85.6% for CG+ compounds, 79.1% for CG− compounds and 83.6% for the overall accuracy. The results indicate that the built models are potentially useful for facilitating the prediction of chemical carcinogenicity of untested compounds.

Identification of small molecule aggregators from large compound libraries by support vector machines

Small molecule aggregators non‐specifically inhibit multiple unrelated proteins, rendering them therapeutically useless. They frequently appear as false hits and thus need to be eliminated in high‐throughput screening campaigns. Computational methods have been explored for identifying aggregators, which have not been tested in screening large compound libraries. We used 1319 aggregators and 128,325 non‐aggregators to develop a support vector machines (SVM) aggregator identification model, which was tested by four methods. The first is five fold cross‐validation, which showed comparable aggregator and significantly improved non‐aggregator identification rates against earlier studies. The second is the independent test of 17 aggregators discovered independently from the training aggregators, 71% of which were correctly identified. The third is retrospective screening of 13M PUBCHEM and 168K MDDR compounds, which predicted 97.9% and 98.7% of the PUBCHEM and MDDR compounds as non‐aggregators. The fourth is retrospective screening of 5527 MDDR compounds similar to the known aggregators, 1.14% of which were predicted as aggregators. SVM showed slightly better overall performance against two other machine learning methods based on five fold cross‐validation studies of the same settings. Molecular features of aggregation, extracted by a feature selection method, are consistent with published profiles. SVM showed substantial capability in identifying aggregators from large libraries at low false‐hit rates.

COMPUTATIONAL METHOD FOR DRUG TARGET SEARCH AND APPLICATION IN DRUG DISCOVERY

Ligand-protein inverse docking has recently been introduced as a computer method for identification of potential protein targets of a drug. A protein structure database is searched to find proteins to which a drug can bind or weakly bind. Examples of potential applications of this method in facilitating drug discovery include: (1) identification of unknown and secondary therapeutic targets of a drug, (2) prediction of potential toxicity and side effect of an investigative drug, and (3) probing molecular mechanism of bioactive herbal compounds such as those extracted from plants used in traditional medicines. This method and recent results on its applications in solving various drug discovery problems are reviewed.

Physicochemical Profiles of the Marketed Agrochemicals and Clues for Agrochemical Lead Discovery and Screening Library Development.

Combinatorial chemistry, high‐throughput and virtual screening technologies have been extensively used for discovering agrochemical leads from chemical libraries. The knowledge of the physicochemical properties of the marketed agrochemicals is useful for guiding the design and selection of such libraries. Since the earlier profiling of marketed agrochemicals, the number and types of marketed agrochemicals have significantly increased. Recent studies have shown the change of some physicochemical properties of oral drugs with time. There is a need to also profile the physicochemical properties of the marketed agrochemicals. In this work, we analyzed the key physicochemical properties of 1751 marketed agrochemicals in comparison with the previously‐analyzed herbicides and insecticides, 106 391 natural products and 57 548 diverse synthetic libraries compounds. Our study revealed the distribution profiles and evolution trend of different types of agrochemicals that in many respects are broadly similar to the reported profiles for oral drugs, with the most marked difference being that agrochemicals have a lower number of hydrogen bond donors. The derived distribution patterns provided the rule of thumb guidelines for selecting potential agrochemical leads and also provided clues for further improving the libraries for agrochemical lead discovery.

Interpretation and application of reaction class transition state theory for accurate calculation of thermokinetic parameters using isodesmic reaction method.

We present a further interpretation of reaction class transition state theory (RC-TST) proposed by Truong et al. for the accurate calculation of rate coefficients for reactions in a class. It is found that the RC-TST can be interpreted through the isodesmic reaction method, which is usually used to calculate reaction enthalpy or enthalpy of formation for a species, and the theory can also be used for the calculation of the reaction barriers and reaction enthalpies for reactions in a class. A correction scheme based on this theory is proposed for the calculation of the reaction barriers and reaction enthalpies for reactions in a class. To validate the scheme, 16 combinations of various ab initio levels with various basis sets are used as the approximate methods and CCSD(T)/CBS method is used as the benchmarking method in this study to calculate the reaction energies and energy barriers for a representative set of five reactions from the reaction class: R(c)CH(R(b))CR(a)CH2 + OH(•) → R(c)C(•)(R(b))CR(a)CH2 + H2O (R(a), R(b), and R(c) in the reaction formula represent the alkyl or hydrogen). Then the results of the approximate methods are corrected by the theory. The maximum values of the average deviations of the energy barrier and the reaction enthalpy are 99.97 kJ/mol and 70.35 kJ/mol, respectively, before correction and are reduced to 4.02 kJ/mol and 8.19 kJ/mol, respectively, after correction, indicating that after correction the results are not sensitive to the level of the ab initio method and the size of the basis set, as they are in the case before correction. Therefore, reaction energies and energy barriers for reactions in a class can be calculated accurately at a relatively low level of ab initio method using our scheme. It is also shown that the rate coefficients for the five representative reactions calculated at the BHandHLYP/6-31G(d,p) level of theory via our scheme are very close to the values calculated at CCSD(T)/CBS level. Finally, reaction barriers and reaction enthalpies and rate coefficients of all the target reactions calculated at the BHandHLYP/6-31G(d,p) level of theory via the same scheme are provided.