Dimensionality of amino acid space and solvent accessibility prediction with neural networks

Solvent accessibility prediction from amino acid sequences has been pursued by several researchers. Such a prediction typically starts by transforming the amino acid category (or type) information into numerical representations. All twenty amino acids can be completely and uniquely represented by 20-dimensional vectors. Here, we investigate if the amino acid space defined in this way really requires twenty dimensions. We tried to develop corresponding representations in fewer dimensions. A method for searching optimal codification schema in an arbitrary space using neural networks was developed. The method is used to obtain optimal encoding of amino acids at various levels of dimensionality, and applied to optimize the amino acid codifications for the prediction of the solvent accessibility values of the proteins using feed-forward neural networks. The traditional 20-dimensional codification seems to be redundant in solving the solvent accessibility prediction problem, since a 1-dimensional codification is able to achieve almost the same degree of accuracy as the 20-dimensional codification. Optimal coding in much fewer dimensions could be used to make the predictions of accessible surface area with almost the same degree of accuracy as that obtained by a fully unique 20-dimensional coding. The 1-dimensional amino acid codification for solvent accessibility prediction obtained by a purely mathematical way based on neural networks is highly correlated with a physical property of the amino acids, namely their average solvent accessibility. The method developed to find the optimal codification is general, although the codification thus produced is dependent on the type of estimated property.     

从药物的三维分子结构预测人体小肠吸收

在口服药物的发展过程中，人体小肠吸收的预测是候选药物设计、优化和选择的一个主要目标.VolSurf/GRID计算方法作为一个新的工具被用来预测被测化合物的人体小肠吸收，以及测定人体小肠吸收所必需的重要的分子特征.被测化合物包括112个结构不同的类似药物的化合物.使用偏最小二乘判别分析方法在实验数据和人体小肠吸收的理论分子特征之间建立相关性.建立的两个模型之间具有较高的一致性.小肠吸收实验数据与分子特征之间好的相关性（r2=0.82, q2=0.67）表明,从化合物的三维分子结构能够预测人体小肠吸收.有利于人体小肠吸收的药物分子特征包括,分子量中心与亲水区重心的不平衡性，较大的疏水区域以及分子内较少的氢键给体.     

Molecular modeling of polymer-clay nanocomposite precursors: Lysine in montmorillonite interlayers

The layered structure of clays with interlayer cations leads to unique chemical and mechanical properties, which have been capitalized on in the field of polymer/layered silicate nanocomposites. Hydrophilic silica surfaces can become organophilic with the inclusion of alkylammonium cations, which improve the wetting characteristics of the polymer matrix. In fact, the molecular level interactions of amino acids, either natural or non-natural, with clay surfaces are at the heart of fields of study as diverse as nanocomposites fabrication, drug delivery, bio-remediation of soils and catalysis of biological polymers, to name a few. The ubiquity of these systems and the potential uses to which they could be put suggests the necessity of a deeper understanding of the interplay of bonds, conformations, and configurations between the molecules and the hosts. The interactions of the amino acid lysine with sodium montmorillonite were studied using theoretical molecular modeling methods. The interlayer spacing of montmorillonite was increased by incorporating water molecules and allowing the system to evolve with molecular mechanics. Care was taken to retain the sodium cations in the interlayer. The initial amino acid conformation was obtained surrounding the molecule with numerous discrete water molecules and minimizing the system at the semi empirical level. The optimized amino acid was then placed in the interlayer space in a series of initial positions. Molecular mechanics calculations were performed and the final positions were analyzed. The results tended to indicate the preponderance of configurations which included surface-sodium-amino acid complexes with a variety of spatial arrangements. These results were compared with molecular dynamics calculations of similar systems from the literature.     

微波促进一锅法合成氨基甲酸酯型α-猪去氧胆酸分子钳

在微波辐射条件下, 以α-猪去氧胆酸为隔离基, 通过三光气桥连各种芳香胺, 以很好的产率合成了一系列新的手性分子钳, 其结构经¹H NMR, IR, MS和元素分析确证, 并且考察了其对中性分子和D/L-氨基酸甲酯的识别性能. 实验结果表明, 这类分子钳人工受体不仅对中性有机小分子具有优良的识别性能, 而且对D/L-氨基酸甲酯亦具有良好的对映选择性识别能力.     

A deep learning approach for the blind logP prediction in SAMPL6 challenge

Water octanol partition coefficient serves as a measure for the lipophilicity of a molecule and is important in the field of drug discovery. A novel method for computational prediction of logarithm of partition coefficient (logP) has been developed using molecular fingerprints and a deep neural network. The machine learning model was trained on a dataset of 12,000 molecules and tested on 2000 molecules. In this article, we present our results for the blind prediction of logP for the SAMPL6 challenge. While the best submission achieved a RMSE of 0.41 logP units, our submission had a RMSE of 0.61 logP units. Overall, we ranked in the top quarter out of the 92 submissions that were made. Our results show that the deep learning model can be used as a fast, accurate and robust method for high throughput prediction of logP of small molecules.     

α-猪去氧胆酸类分子钳的设计合成

以α-猪去氧胆酸为隔离基，芳香族化合物为手臂设计合成了一类新型的酯键 型分子钳，结构均经^1H NMR，IR,MS及元素分析所确证，并且考察了其对芳香胺类 化合物和D／L-氨基酸甲酯的识别性能．初步研究结果表明，这类分子钳受体不但 对中性小分子具有识别性能，而且对D／L-氨基酸甲酯具有对映选择性识别性能．     

Prediction of n-octanol/water partition coefficients from PHYSPROP database using artificial neural networks and E-state indices

A new method, ALOGPS v 2.0 (http://www.lnh.unil.ch/~itetko/logp/), for the assessment of n-octanol/water partition coefficient, log P, was developed on the basis of neural network ensemble analysis of 12 908 organic compounds available from PHYSPROP database of Syracuse Research Corporation. The atom and bond-type E-state indices as well as the number of hydrogen and non-hydrogen atoms were used to represent the molecular structures. A preliminary selection of indices was performed by multiple linear regression analysis, and 75 input parameters were chosen. Some of the parameters combined several atom-type or bond-type indices with similar physicochemical properties. The neural network ensemble training was performed by efficient partition algorithm developed by the authors. The ensemble contained 50 neural networks, and each neural network had 10 neurons in one hidden layer. The prediction ability of the developed approach was estimated using both leave-one-out (LOO) technique and training/test protocol. In case of interseries predictions, i.e., when molecules in the test and in the training subsets were selected by chance from the same set of compounds, both approaches provided similar results. ALOGPS performance was significantly better than the results obtained by other tested methods. For a subset of 12 777 molecules the LOO results, namely correlation coefficient r(2)= 0.95, root mean squared error, RMSE = 0.39, and an absolute mean error, MAE = 0.29, were calculated. For two cross-series predictions, i.e., when molecules in the training and in the test sets belong to different series of compounds, all analyzed methods performed less efficiently. The decrease in the performance could be explained by a different diversity of molecules in the training and in the test sets. However, even for such difficult cases the ALOGPS method provided better prediction ability than the other tested methods. We have shown that the diversity of the training sets rather than the design of the methods is the main factor determining their prediction ability for new data. A comparative performance of the methods as well as a dependence on the number of non-hydrogen atoms in a molecule is also presented.     

Genius: a genetic algorithm for automated structure elucidation from 13C NMR spectra

The automated structure elucidation of organic molecules from experimentally obtained properties is extended by an entirely new approach. A genetic algorithm is implemented that uses molecular constitution structures as individuals. With this approach, the structure of organic molecules can be optimized to meet experimental criteria, if in addition a fast and accurate method for the prediction of the used physical or chemical features is available. This is demonstrated using 13C NMR spectrum as readily obtainable information. By means of artificial neural networks a fast and accurate method for calculating the 13C NMR spectrum of the generated structures exists. The method is implemented and tested successfully for organic molecules with up to 18 non-hydrogen atoms.     

海藻糖和氨基酸之间相互作用的分子动力学模拟

虽然海藻糖已经广泛用于蛋白质稳定性研究,但海藻糖稳定蛋白质的作用机理尚不清晰.本文利用全原子分子动力学模拟研究了20种常见氨基酸和海藻糖之间的分子机理.结果表明,所有氨基酸,尤其是极性和带电氨基酸,均优先与水分子结合.相反,仅有疏水性氨基酸与海藻糖发生相互作用,尤其是芳香族和疏水性氨基酸的侧链更易于和海藻糖接触.所有氨基酸的主链与水分子接触的趋势一致.虽然氨基酸和海藻糖与水之间均形成氢键,但氨基酸和海藻糖之间的氢键相互作用要弱于氨基酸和水之间的氢键相互作用.上述分子模拟的结果对于海藻糖稳定蛋白质作用机理的解析及高效蛋白质稳定剂的理性设计具有非常重要的理论指导意义.     

海藻糖和氨基酸之间相互作用的分子动力学模拟

虽然海藻糖已经广泛用于蛋白质稳定性研究,但海藻糖稳定蛋白质的作用机理尚不清晰. 本文利用全原子分子动力学模拟研究了20种常见氨基酸和海藻糖之间的分子机理. 结果表明,所有氨基酸,尤其是极性和带电氨基酸,均优先与水分子结合. 相反,仅有疏水性氨基酸与海藻糖发生相互作用,尤其是芳香族和疏水性氨基酸的侧链更易于和海藻糖接触. 所有氨基酸的主链与水分子接触的趋势一致. 虽然氨基酸和海藻糖与水之间均形成氢键,但氨基酸和海藻糖之间的氢键相互作用要弱于氨基酸和水之间的氢键相互作用. 上述分子模拟的结果对于海藻糖稳定蛋白质作用机理的解析及高效蛋白质稳定剂的理性设计具有非常重要的理论指导意义.     

Estimation of molecular similarity based on 4D-QSAR analysis: formalism and validation

The 4D-QSAR paradigm has been used to develop a formalism to estimate molecular similarity measures as a function of conformation, alignment, and atom type. It is possible to estimate the molecular similarity of pairs of molecules based upon the entire ensemble of conformational states each molecule can adopt at a given temperature, normally room temperature. Molecular similarity can be measured in terms of the types of atoms composing each molecule leading to multiple measures of molecular similarity. Multiple measures of molecular similarity can also arise from using different alignment rules to perform relative molecular similarity, RMS, analysis. An alignment independent method of determining molecular similarity measures, referred to as absolute molecular similarity, AMS, analysis has been developed. Various sets and libraries of compounds, including the amino acids, are analyzed using 4D-QSAR molecular similarity analysis to demonstrate the features of the formalism. Exploration of molecular similarity as a function of chirality, identification of common embedded 3D pharmacophores in compounds, and elucidation of 3D-isosteric compounds from structurally diverse libraries are carried out in the application studies.     

Identification of critical chemical features for Aurora kinase-B inhibitors using Hip-Hop, virtual screening and molecular docking

This study was performed to find the selective chemical features for Aurora kinase-B inhibitors using the potent methods like Hip-Hop, virtual screening, homology modeling, molecular dynamics and docking. The best hypothesis, Hypo1 was validated toward a wide range of test set containing the selective inhibitors of Aurora kinase-B. Homology modeling and molecular dynamics studies were carried out to perform the molecular docking studies. The best hypothesis Hypo1 was used as a 3D query to screen the chemical databases. The screened molecules from the databases were sorted based on ADME and drug like properties. The selective hit compounds were docked and the hydrogen bond interactions with the critical amino acids present in Aurora kinase-B were compared with the chemical features present in the Hypo1. Finally, we suggest that the chemical features present in the Hypo1 are vital for a molecule to inhibit the Aurora kinase-B activity.     

QSAR Study and VolSurf Characterization of Human Intestinal Absorption of Druge

The prediction of human intestinal absorption is a major goal in the design,optimization,and selection of candidates for the develoment of oral drugs.In this study,a computerized method(VolSurf with GRID) was used as a novel tool for predicting human intestinal absorption of test compound,and for determining the critical molecular properties needed for human intestinal absorption.The tested molecules consisted of 20 diverse drug-like compounds.Partial least squares(PLS) discriminant analysis was used to correlate the experimental data with the theoretical molecular properties of human intestinal absorption.A good correlation(r^2=0.95,q^2=0.86) between the molecular modeling results and the experimental data demonstrated that human intestinal absorption could be predicted from the three-dimensional(3D) molecular structure of a compound .Favorable structureal properties identified for the potent intestinal absorption of drugs included strong imbalance between the center of mass of a molecule and the barycentre of its hydrophilic and hydrophobic regions and a definitive hydrophobic region as well as less hydrogen bonding donors and acceptors in the molecule.