Mangiferin, an anti-HIV-1 agent targeting protease and effective against resistant strains

The anti-HIV-1 activity of mangiferin was evaluated. Mangiferin can inhibit HIV-1(Ⅲ)(B) induced syncytium formation at non-cytotoxic concentrations, with a 50% effective concentration (EC??) at 16.90 μM and a therapeutic index (TI) above 140. Mangiferin also showed good activities in other laboratory-derived strains, clinically isolated strains and resistant HIV-1 strains. Mechanism studies revealed that mangiferin might inhibit the HIV-1 protease, but is still effective against HIV peptidic protease inhibitor resistant strains. A combination of docking and pharmacophore methods clarified possible binding modes of mangiferin in the HIV-1 protease. The pharmacophore model of mangiferin consists of two hydrogen bond donors and two hydrogen bond acceptors. Compared to pharmacophore features found in commercially available drugs, three pharmacophoric elements matched well and one novel pharmacophore element was observed. Moreover, molecular docking analysis demonstrated that the pharmacophoric elements play important roles in binding HIV-1 protease. Mangiferin is a novel nonpeptidic protease inhibitor with an original structure that represents an effective drug development strategy for combating drug resistance.     

HIV-1 protease cleavage site prediction based on amino acid property

Knowledge of the polyprotein cleavage sites by HIV protease will refine our understanding of its specificity, and the information thus acquired is useful for designing specific and efficient HIV protease inhibitors. Recently, several works have approached the HIV-1 protease specificity problem by applying a number of classifier creation and combination methods. The pace in searching for the proper inhibitors of HIV protease will be greatly expedited if one can find an accurate, robust, and rapid method for predicting the cleavage sites in proteins by HIV protease. In this article, we selected HIV-1 protease as the subject of the study. 299 oligopeptides were chosen for the training set, while the other 63 oligopeptides were taken as a test set. The peptides are represented by features constructed by AAIndex (Kawashima et al., Nucleic Acids Res 1999, 27, 368; Kawashima and Kanehisa, Nucleic Acids Res 2000, 28, 374). The mRMR method (Maximum Relevance, Minimum Redundancy; Ding and Peng, Proc Second IEEE Comput Syst Bioinformatics Conf 2003, 523; Peng et al., IEEE Trans Pattern Anal Mach Intell 2005, 27, 1226) combining with incremental feature selection (IFS) and feature forward search (FFS) are applied to find the two important cleavage sites and to select 364 important biochemistry features by jackknife test. Using KNN (K-nearest neighbors) to combine the selected features, the prediction model obtains high accuracy rate of 91.3% for Jackknife cross-validation test and 87.3% for independent-set test. It is expected that our feature selection scheme can be referred to as a useful assistant technique for finding effective inhibitors of HIV protease, especially for the scientists in this field.     

自适应模糊偏最小二乘方法在药物构效关系建模中的应用

作为一种局部逼近方法,自适应神经模糊推理系统(ANFIS)适于为药物定量构效关系(QSAR)建模。描述药物分子结构的参数较多,常存在耦合关系,会增加建模难度,并影响模型的预报性能。为此,将ANFIS和偏最小二乘(PLS)相结合,先由PLS从样本数据中提取成分,再由ANFIS实现每对成分间的非线性映射,并基于输出误差进一步修正所提取的成分,使之对因变量具有最优的解释能力,由此构建为EB-AFPLS方法。该法已成功地应用于HIV-1蛋白酶抑制剂的QSAR建模,效果良好,显示出很强的学习能力,所建模型的预报性能也优于其它方法。     

Cellular automata and its applications in protein bioinformatics

With the explosion of protein sequences generated in the postgenomic era, it is highly desirable to develop high-throughput tools for rapidly and reliably identifying various attributes of uncharacterized proteins based on their sequence information alone. The knowledge thus obtained can help us timely utilize these newly found protein sequences for both basic research and drug discovery. Many bioinformatics tools have been developed by means of machine learning methods. This review is focused on the applications of a new kind of science (cellular automata) in protein bioinformatics. A cellular automaton (CA) is an open, flexible and discrete dynamic model that holds enormous potentials in modeling complex systems, in spite of the simplicity of the model itself. Researchers, scientists and practitioners from different fields have utilized cellular automata for visualizing protein sequences, investigating their evolution processes, and predicting their various attributes. Owing to its impressive power, intuitiveness and relative simplicity, the CA approach has great potential for use as a tool for bioinformatics.     

Support Vector Machines for predicting HIV protease cleavage sites in protein

Knowledge of the polyprotein cleavage sites by HIV protease will refine our understanding of its specificity, and the information thus acquired is useful for designing specific and efficient HIV protease inhibitors. The pace in searching for the proper inhibitors of HIV protease will be greatly expedited if one can find an accurate, robust, and rapid method for predicting the cleavage sites in proteins by HIV protease. In this article, a Support Vector Machine is applied to predict the cleavability of oligopeptides by proteases with multiple and extended specificity subsites. We selected HIV-1 protease as the subject of the study. Two hundred ninety-nine oligopeptides were chosen for the training set, while the other 63 oligopeptides were taken as a test set. Because of its high rate of self-consistency (299/299 = 100%), a good result in the jackknife test (286/299 = 95%) and correct prediction rate (55/63 = 87%), it is expected that the Support Vector Machine method can be referred to as a useful assistant technique for finding effective inhibitors of HIV protease, which is one of the targets in designing potential drugs against AIDS. The principle of the Support Vector Machine method can also be applied to analyzing the specificity of other multisubsite enzymes.     

Rational design and synthesis of a novel class of active site-targeted HIV protease inhibitors containing a hydroxymethylcarbonyl isostere. Use of phenylnorstatine or allophenylnorstatine as a transition-state mimic.

A novel class of HIV-1 protease inhibitors containing a hydroxymethylcarbonyl (HMC) isostere were designed from the substrate transition state and synthesized. Phenylnorstatine [Pns; (2R,3S)-3-amino-2-hydroxy-4-phenylbutyric acid] and the 2S diastereomer, (2S,3S)-3-amino-2-hydroxy-4-phenylbutyric acid, named allophenylnorstatine (Apns) were effective transition-state mimics, and incorporation of Pns-Pro or Apns-Pro at the P1-P1' site gave potent and specific HIV-1 protease inhibitors. In the inhibitory assays, the chemically synthesized [Ala67,95] HIV-1 protease was used.     

Automated derivation of reaction rules for the EROS 6.0 system for reaction prediction

The purpose of the EROS 6.0 system is to predict the products of chemical reactions and to model reaction mechanisms. This is accomplished by elementary reaction steps that are selected through the rules that constitute the knowledge base of the EROS 6.0 system. These rules are derived by methods of machine learning. The learning process is based on reaction in data bases. An overview of the EROS 6.0 system is given and the structure of the knowledge as well as the generation of reaction rules are described.     

Individually double minimum-distance definition of protein–RNA binding residues and application to structure-based prediction

Identifying protein–RNA binding residues is essential for understanding the mechanism of protein–RNA interactions. So far, rigid distance thresholds are commonly used to define protein–RNA binding residues. However, after investigating 182 non-redundant protein–RNA complexes, we find that it would be unsuitable for a certain amount of complexes since the distances between proteins and RNAs vary widely. In this work, a novel definition method was proposed based on a flexible distance cutoff. This method can fully consider the individual differences among complexes by setting a variable tolerance limit of protein–RNA interactions, i.e. the double minimum-distance by which different distance thresholds are achieved for different complexes. In order to validate our method, a comprehensive comparison between our flexible method and traditional rigid methods was implemented in terms of interface structure, amino acid composition, interface area and interaction force, etc. The results indicate that this method is more reasonable because it incorporates the specificity of different complexes by extracting the important residues lost by rigid distance methods and discarding some redundant residues. Finally, to further test our double minimum-distance definition strategy, we developed a classifier to predict those binding sites derived from our new method by using structural features and a random forest machine learning algorithm. The model achieved a satisfactory prediction performance and the accuracy on independent data sets reaches to 85.0%. To the best of our knowledge, it is the first prediction model to define positive and negative samples using a flexible cutoff. So the comparison analysis and modeling results have demonstrated that our method would be a very promising strategy for more precisely defining protein–RNA binding sites.     

Theoretical study on the mechanism of a ring-opening reaction of oxirane by the active-site aspartic dyad of HIV-1 protease

Two possible mechanisms of the irreversible inhibition of HIV-1 protease by epoxide inhibitors are investigated on an enzymatic model using ab initio (MP2) and density functional theory (DFT) methods (B3LYP, MPW1K and M05-2X). The calculations predict the inhibition as a general acid-catalyzed nucleophilic substitution reaction proceeding by a concerted SN2 mechanism with a reaction barrier of ca. 15-21 kcal mol(-1). The irreversible nature of the inhibition is characterized by a large negative reaction energy of ca. -17-(-24) kcal mol(-1). A mechanism with a direct proton transfer from an aspartic acid residue of the active site onto the epoxide ring has been shown to be preferred compared to one with the proton transfer from the acid catalyst facilitated by a bridging catalytic water molecule. Based on the geometry of the transition state, structural data important for the design of irreversible epoxide inhibitors of HIV-1 protease were defined. Here we also briefly discuss differences between the epoxide ring-opening reaction in HIV-1 protease and epoxide hydrolase, and the accuracy of the DFT method used.     

The effect of physical organic properties on hydrophobic fields

Summary Physical organic structural properties of small molecules and macromolecules such as bond count, branching and proximity between multiple polar fragments contribute significantly to measured hydrophobicity (log P). These structural properties are encoded in the Rekker and Leo methods of calculating log P as structural-dependent factors. Regardless of the size of the atom primitive set, methods predicting log P with only atom primitives can miss subtle structural detail within series of related compounds. The HINT (Hydropathic INTeractions) model for inter- and intramolecular noncovalent interactions calculates atom-based hydrophobic constants, but uses all Leo-type factors in the calculation rather than a large set of atom primitives. Two types of applications of HINT are discussed: evaluation of the binding of an inhibitor (A74704) to HIV-1 protease, where it is shown that modeling of the protonation state (i.e., Asp²⁵, Asp¹²⁵) in the protein can strongly influence perceived substrate binding; and the use of HINT to calculate a third (hydropathic) field for CoMFA can yield a statistically enhanced and predictive model for molecular design.     

CRNNTL: Convolutional Recurrent Neural Network and Transfer Learning for QSAR Modeling in Organic Drug and Material Discovery

Molecular latent representations, derived from autoencoders (AEs), have been widely used for drug or material discovery over the past couple of years. In particular, a variety of machine learning methods based on latent representations have shown excellent performance on quantitative structure–activity relationship (QSAR) modeling. However, the sequence feature of them has not been considered in most cases. In addition, data scarcity is still the main obstacle for deep learning strategies, especially for bioactivity datasets. In this study, we propose the convolutional recurrent neural network and transfer learning (CRNNTL) method inspired by the applications of polyphonic sound detection and electrocardiogram classification. Our model takes advantage of both convolutional and recurrent neural networks for feature extraction, as well as the data augmentation method. According to QSAR modeling on 27 datasets, CRNNTL can outperform or compete with state-of-art methods in both drug and material properties. In addition, the performances on one isomers-based dataset indicate that its excellent performance results from the improved ability in global feature extraction when the ability of the local one is maintained. Then, the transfer learning results show that CRNNTL can overcome data scarcity when choosing relative source datasets. Finally, the high versatility of our model is shown by using different latent representations as inputs from other types of AEs.     

Efficient Stochastic Simulation Algorithm for Chemically Reacting Systems Based on Support Vector Regression

The stochastic simulation algorithm （SSA） accurately depicts spatially homogeneous wellstirred chemically reacting systems with small populations of chemical species and properly represents noise, but it is often abandoned when modeling larger systems because of its computational complexity. In this work, a twin support vector regression based stochastic simulations algorithm （TS^3A） is proposed by combining the twin support vector regression and SSA, the former is a well-known robust regression method in machine learning. Numerical results indicate that this proposed algorithm can be applied to a wide range of chemically reacting systems and obtain significant improvements on efficiency and accuracy with fewer simulating runs over the existing methods.     

Machine learning-, rule- and pharmacophore-based classification on the inhibition of P-glycoprotein and NorA

The efflux pumps P-glycoprotein (P-gp) in humans and NorA in Staphylococcus aureus are of great interest for medicinal chemists because of their important roles in multidrug resistance (MDR). The high polyspecificity as well as the unavailability of high-resolution X-ray crystal structures of these transmembrane proteins lead us to combining ligand-based approaches, which in the case of this study were machine learning, perceptual mapping and pharmacophore modelling. For P-gp inhibitory activity, individual models were developed using different machine learning algorithms and subsequently combined into an ensemble model which showed a good discrimination between inhibitors and noninhibitors (acc_{train-diverse} = 84%; acc_{internal-test} = 92% and acc_{external-test} = 100%). For ligand promiscuity between P-gp and NorA, perceptual maps and pharmacophore models were generated for the detection of rules and features. Based on these in silico tools, hit compounds for reversing MDR were discovered from the in-house and DrugBank databases through virtual screening in an attempt to restore drug sensitivity in cancer cells and bacteria.     

An ensemble learning framework for potential miRNA-disease association prediction with positive-unlabeled data

To explore the pathogenic mechanisms of MicroRNA (miRNA) on diverse diseases, many researchers have concentrated on discovering the potential associations between miRNA and disease using machine learning methods. However, the prediction accuracy of supervised machine learning methods is limited by lacking of experimentally-validated uncorrelated miRNA-disease pairs. Without these negative samples, training a highly accurate model is much more difficult. Different from traditional miRNA-disease prediction models using randomly selected unknown samples as negative training samples, we propose an ensemble learning framework to solve this positive-unlabeled (PU) learning problem. The framework incorporates two steps, i.e., a novel semi-supervised Kmeans (SS-Kmeans) to extract reliable negative samples from unknown miRNA-disease pairs and subagging method to generate diverse training sample sets to make full use of those reliable negative samples for ensemble learning. Combined with effective random vector functional link (RVFL) network as prediction model, the proposed framework showed superior prediction accuracy comparing with other popular approaches. A case study on lung and gastric neoplasms further confirms the framework’s efficacy at identifying miRNA disease associations.     

Computational studies of darunavir into HIV-1 protease and DMPC bilayer: necessary conditions for effective binding and the role of the flaps

Human immunodeficiency virus type 1 protease (HIV-1 PR) is one of the main targets toward AIDS therapy. We have selected the potent drug darunavir and a weak inhibitor (fullerene analog) as HIV-1 PR substrates to compare protease's conformational features upon binding. Molecular dynamics (MD), molecular mechanics Poisson-Boltzmann surface area (MM-PBSA), and quantum-mechanical (QM) calculations indicated the importance of the stability of HIV-1 PR flaps toward effective binding: a weak inhibitor may induce flexibility to the flaps, which convert between closed and semiopen states. A water molecule in the darunavir-HIV-1 PR complex bridged the two flap tips of the protease through hydrogen bonding (HB) interactions in a stable structure, a feature that was not observed for the fullerene-HIV-1 PR complex. Additionally, despite that van der Waals interactions and nonpolar contribution to solvation favored permanent fullerene entrapment into the cavity, these interactions alone were not sufficient for effective binding; enhanced electrostatic interactions as observed in the darunavir-complex were the crucial component of the binding energy. An alternative pathway to the usual way of a ligand to access the cavity was also observed for both compounds. Each ligand entered the binding cavity through an opening between the one flap of the protease and a neighboring loop. This suggested that access to the cavity is not necessarily regulated by flap opening. Darunavir exerts its biological action inside the cell, after crossing the membrane barrier. Thus, we also initiated a study on the interactions between darunavir and the DMPC bilayer to reveal that the drug was accommodated inside the bilayer in conformations that resembled its structure into HIV-1 PR, being stabilized via HBs with the lipids and water molecules.     

Potent HIV-1 protease inhibitors incorporating meso-bicyclic urethanes as P2-ligands: structure-based design, synthesis, biological evaluation and protein-ligand X-ray studies

Recently, we designed a series of novel HIV-1 protease inhibitors incorporating a stereochemically defined bicyclic fused cyclopentyl (Cp-THF) urethane as the high affinity P2-ligand. Inhibitor with this P2-ligand has shown very impressive potency against multi-drug-resistant clinical isolates. Based upon the -bound HIV-1 protease X-ray structure, we have now designed and synthesized a number of meso-bicyclic ligands which can conceivably interact similarly to the Cp-THF ligand. The design of meso-ligands is quite attractive as they do not contain any stereocenters. Inhibitors incorporating urethanes of bicyclic-1,3-dioxolane and bicyclic-1,4-dioxane have shown potent enzyme inhibitory and antiviral activities. Inhibitor (K(i) = 0.11 nM; IC(50) = 3.8 nM) displayed very potent antiviral activity in this series. While inhibitor showed comparable enzyme inhibitory activity (K(i) = 0.18 nM) its antiviral activity (IC(50) = 170 nM) was significantly weaker than inhibitor . Inhibitor maintained an antiviral potency against a series of multi-drug resistant clinical isolates comparable to amprenavir. A protein-ligand X-ray structure of -bound HIV-1 protease revealed a number of key hydrogen bonding interactions at the S2-subsite. We have created an active model of inhibitor based upon this X-ray structure.