Monodisperse manganese oxide flowerlike nanostructures have been prepared facilely at low temperature and ambient atmosphere. The effect of the reaction time on the microstructure and morphology is observed systemically by transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM). Meanwhile, the possible formation mechanism of the flowerlike nanostructures has been proposed and discussed. It is also found that the reaction temperature has great influences on the morphology of these unique nanostructures. The results of nitrogen adsorption−desorption experiments and electrochemical measurements show that the product obtained at 40 °C for 8 h has large specific surface area, uniform pore size distribution, and excellent capacitance performance, which make it a potential supercapacitor electrode material.

•Data mining methods were used to establish the correlation about the specific surface area (SSA) with five features including two atomic parameters and three technical parameters.•The layered double hydroxide (LDH) of Ni-Fe-CO3 system with desired specific surface area was controllably synthesized on the basis of model prediction rather than try and error method.•Useful hints for the discovery of novel materials with the assistance of machine learning were proved. The time to design LDH with desired properties will be significantly reduced.The value of specific surface area (SSA) of layered double hydroxide (LDH) can be employed as the main parameters of adsorbent evaluation. In this work, data mining methods were used to explore the correlations of the SSA (ranged 10–90 m2 g−1) with their chemical compositions and technical parameters in search of the certain LDH material with desired SSA. The genetic algorithm (GA)-support vector regression (SVR) method was used to filter the main molecular descriptors for modeling. The related coefficient (R) between predicted SSA and experimental SSA reached as high as 0.937 for training set and 0.892 for leave-one-out cross validation (LOOCV), respectively. The optimal model was then applied to 9 independent samples to test the prediction ability with the mean relative error equal to 14.7%. A case study of controllable synthesis predicted by the model was also carried out, and the new LDH material (Ni-Fe CO3 LDH) with desired SSA was verified by our experiments with the relative error equal to 13.8%. The method outlined here can provide valuable hints into the exploration of materials design with the assistance of machine learning.Download high-res image (195KB)Download full-size image

•Both qualitative and quantitative methods adopted widely in materials data mining (MDM) have been systematically reviewed to meet different tasks of materials discovery and optimization.•The novel qualitative method by using optimal projection recognition technique is reviewed in detail for controllable synthesis of dendritic Co3O4 superstructures based on pattern recognition classification diagram.•The detailed MDM process has been demonstrated in case study on materials design of layered double hydroxide with desired basal spacing based on the quantitative modelling method called relevance vector machine.•The state-of-the-arts of data mining-aided battery materials discovery and thermoelectric materials design have been reviewed, indicating that MDM approach may play a more important role to discover novel materials in future.Recent developments in data mining-aided materials discovery and optimization are reviewed in this paper, and an introduction to the materials data mining (MDM) process is provided using case studies. Both qualitative and quantitative methods in machine learning can be adopted in the MDM process to accomplish different tasks in materials discovery, design, and optimization. State-of-the-art techniques in data mining-aided materials discovery and optimization are demonstrated by reviewing the controllable synthesis of dendritic Co3O4 superstructures, materials design of layered double hydroxide, battery materials discovery, and thermoelectric materials design. The results of the case studies indicate that MDM is a powerful approach for use in materials discovery and innovation, and will play an important role in the development of the Materials Genome Initiative and Materials Informatics.Download high-res image (348KB)Download full-size image

•The QSPR model of basal spacing of LDH have been built, using the novel machine learning algorithm: Relevance Vector Machine.•The performance of Model is well, either in training data or validation data.•The Ni-Al-CO3, which is selected by High Throughput Virtual Screening, have been synthesized and the prediction is highly consistent with experiment result.•For the researcher who is interested in the basal spacing of LDH, we provide the model as web server (using web browser can access the server), which is helpful to screen out new LDHs with desired basal spacing.A Quantitative Structure Property Relationship (QSPR) model for the basal spacing of layered double hydroxide is developed in the present work by using generic algorithms feature selection, and relevance vector machine regression. The relative error of the developed model is 0.78% in cross-validation. Then, the QSPR model is applied to recommend a LDH with desired basal spacing for synthesis. The synthesized LDH meets the design requirement, thereby confirming the prediction power of the developed model.Download high-res image (167KB)Download full-size image

Histone methyltransferases are involved in various biological functions, and these methylation regulating enzymes’ abnormal expression or activity has been noted in several human cancers. Within this context, SET domain-containing (lysine methyltransferase) 7 (SET7, also called KMT7, SETD7, SET9) is of increasing significance due to its diverse roles in biological functions and diseases, such as diabetes, cancers, alopecia areata, atherosclerotic vascular disease, HIV, and HCV. In this study, DC-S100, which was discovered by pharmacophore- and docking-based virtual screening, was identified as the hit compound of SET7 inhibitor. Structure–activity relationship (SAR) analysis was performed on analogs of DC-S100 and according to the putative binding mode of DC-S100, structure modifications were made to improve its activity. Of note, compounds DC-S238 and DC-S239, with IC50 values of 4.88 and 4.59 μM, respectively, displayed selectivity for DNMT1, DOT1L, EZH2, NSD1, SETD8, and G9a. Taken together, DC-S238 and DC-S239 can serve as leads for further investigation as SET7 inhibitors and the chemical toolkits for functional biology studies of SET7.

•Mesoporous Co–Al LDHs with rodlike and nanosheet-like morphologies were synthesized via micro-emulsion method.•The calcined Co–Al sample with 10% content of cobalt exhibits the highest specific surface area.•The calcined samples exhibit higher fluoride removal efficiency than the uncalcined ones.•Performance of the adsorption decreases with an increase in the cobalt ratio generally.•The adsorption kinetics and isotherms were also discussed.Facile micro-emulsion methods have been developed to synthesize mesoporous Co–Al hydroxide carbonates with rodlike and hexagonal sheetlike morphologies. A series of samples with different molar ratio of cobalt/aluminum were prepared to investigate the impact of cobalt content. Employing transmission electron microscopy, X-ray diffraction analysis and Fourier transformed infrared spectra, the morphology, structure and composition of the products were investigated. Characterized by gas-sorption measurements, the specific surface area of the calcined S-10 sample with the rodlike morphology was as high as 379 m2/g, which provided this material porosity properties and outstanding fluoride removal efficiency as high as 95.6%. The sorptive removal capacity of fluoride from aqueous solutions by the precursors and their corresponding annealed products was investigated in a batch mode. Also, the adsorption kinetics and isotherms were discussed.

•The QSPR model was constructed to relate the atomic parameters of LDHs compounds to their basal spacings.•A new LDH of Mg-Al-CO3 system with the desired basal spacing 7.6 Å was synthesized in our lab.•The desired compound was confirmed by our experiment with the relative error equal to 0.93%.•The QSPR model is used to predict the basal spacing of LDHs compound for the first time as far as we know.Efficient and effective prediction of the basal spacing is of great importance to materials design of layered double hydroxides (LDHs). In this work, the QSPR model was constructed to predict the basal spacing of LDHs from 7.5 to 8.0 Å by using the support vector regression (SVR) algorithm. The genetic algorithm (GA)–support vector regression (SVR) method was used to filter the main molecular descriptors in modeling. The QSPR model available was tested by an external test set consisting of 8 compounds. As a case study of controllable synthesis based on the QSPR model, the new LDH of Mg–Al–CO3 system with the desired basal spacing 7.6 Å, which was screened out from a list of LDH dataset consisting of 30 different kinds of samples, was verified by our experiment with the relative error equal to 0.93%. The method outlined here can be served as a new computational template for the materials design and control synthesis of the LDH with the desired basal spacing based on QSPR model for the first time.

•A model of drug–enzyme interaction with high prediction accuracies is built.•An optimal subset with fewer features is constructed.•Geometry features are the most important of all the features.•Drugs take a leading role in drug and enzyme's binding.It is important to correctly and efficiently map drugs and enzymes to their possible interaction network in modern drug research. In this work, a novel approach was introduced to encode drug and enzyme molecules with physicochemical molecular descriptors and pseudo amino acid composition, respectively. Based on this encoding method, Random Forest was adopted to build the drug–enzyme interaction network. After selecting the optimal features that are able to represent the main factors of drug–enzyme interaction in our prediction, a total of 129 features were attained which can be clustered into nine categories: Elemental Analysis, Geometry, Chemistry, Amino Acid Composition, Secondary Structure, Polarity, Molecular Volume, Codon Diversity and Electrostatic Charge. It is further found that Geometry features were the most important of all the features. As a result, our predicting model achieved an MCC of 0.915 and a sensitivity of 87.9% at the specificity level of 99.8% for 10-fold cross-validation test, and achieved an MCC of 0.895 and a sensitivity of 95.7% at the specificity level of 95.4% for independent set test. This article is part of a Special Issue entitled: Computational Proteomics, Systems Biology & Clinical Implications. Guest Editor: Yudong Cai.

Core–shell Al2O3-supported nickel catalyst was prepared and the products were tested in fixed-bed reactor for biomass tar steam reforming. In our work, toluene is treated as tar destruction model compound. The products were characterized by the means of H2-TPR, XRD, BET, TEM and SEM. Comparing with the nanoparticles alumina-supported counterparts, the catalyst showed not only superior activity, more yields of outgases but also better stability. The stable shells provide the unique environment around active sites and the strong interaction between Ni and the core–shell supports seems to be responsible for the catalyst activity and stability in toluene steam reforming. The methods presented in this work can be generalized in materials design and control synthesis of other nano materials.

Monoclinic phase BiPO4 nanocrystals have been synthesized in the oleic acid (OA) solution through one-pot solvothermal route. The photocatalytic performances of the BiPO4 nanocrystals were evaluated by the degradation of rhodamine B (RhB) solution under UV and visible light irradiation. The reaction rate of BiPO4 nanocrystals before calcination was 0.0509 min−1, which was much higher than that after calcination (0.0328 min−1). The BiPO4 nanocrystals before calcination exhibit higher photocatalytic activity than that after calcination. The small size of BiPO4 nanocrystals may be the key factor for its high photocatalytic performances.

Well-defined porous Co3O4 nanodendrites have been synthesized by a simple one-pot hydrothermal method combined with subsequent calcination. Importantly, after thermal treatment, the dendritic morphology could be completely preserved. The as-obtained superstructures are characterized by several techniques, such as powder X-ray diffraction, Fourier transform infrared spectroscopy, X-ray photoemission spectroscopy, elemental analysis, transmission electron microscopy, high-resolution TEM and magnetometry. On the dendritic hierarchical structures, a number of nanorods with different lengths and widths are connected to the main trunk with a diameter of about 50 nm and length of several micrometers. Each branch is about 0.5–1.0 μm with a width ranging from 100 to 400 nm. The possible formation mechanism was proposed on the basis of the contrasting experiments. It demonstrated that trisodium citrate and acetone play important roles in the synthesis of Co3O4 nanodendrites, while the other reaction parameters, such as reaction temperature, concentration of trisodium citrate and reaction time, have close relationships with the final morphology of the Co3O4 products. Optical properties of Co3O4 nanodendrites were characterized by Raman and UV-vis spectroscopy. Magnetic property measurement shows that Co3O4 nanodendrites have a low Néel transition temperature of 35 K. The as-prepared mesoporous Co3O4 nanostructures are expected to have great applications in many fields.

Materials design is the most important and fundamental work on the background of materials genome initiative for global competitiveness proposed by the National Science and Technology Council of America. As far as the methodologies of materials design, besides the thermodynamic and kinetic methods combing databases, both deductive approaches so-called the first principle methods and inductive approaches based on data mining methods are gaining great progress because of their successful applications in materials design. In this paper, support vector machine (SVM), including support vector classification (SVC) and support vector regression (SVR) based on the statistical learning theory (SLT) proposed by Vapnik, is introduced as a relatively new data mining method to meet the different tasks of materials design in our lab. The advantage of using SVM for materials design is discussed based on the applications in the formability of perovskite or BaNiO3 structure, the prediction of energy gaps of binary compounds, the prediction of sintered cold modulus of sialon-corundum castable, the optimization of electric resistances of VPTC semiconductors and the thickness control of In2O3 semiconductor film preparation. The results presented indicate that SVM is an effective modeling tool for the small sizes of sample sets with great potential applications in materials design.

A large series of pyrrolidine amides derivatives as DPP-IV inhibitors was subjected to quantitative structure–activity relationship (QSAR) analysis. These 248 geometrical structures were constructed and optimized at the HF/6-31G* level of theory by the Gaussian program. The 2D–QSAR model was developed from a training set consisting of 186 compounds by the minimum redundancy maximum relevance–sequential floating back–support vector regression method with a good determination coefficient: the squared correlation coefficient (Rtrain2 = 0.867) and the tenfold cross-validation squared correlation coefficient (qtrain-CV2 = 0.669). The QSAR model was then tested using an external test set consisting of 62 compounds and provided a satisfactory external predictive ability (Rtest2 = 0.666). 2D–QSAR model is robust and reliable when compared with 3D–QSAR techniques for the analogous compounds. According to the QSAR analysis, the electronic effect plays an important role for the substituents of the pyrrolidine and carbon rings. The study would serve as a guideline in designing more potent and selective drugs against type 2 diabetes.

A novel screw-cap-like indium tin oxide (ITO) precursor and the corresponding porous ITO particle are successfully achieved by a facile shape-controlled synthesis at room temperature. The screw-cap-like ITO precursors show a unique hexagonal structure with a hole in the center, which may stem from self-assembly of precursor nanosheets and following chemical etching. The corresponding porous screw-cap-like ITO, composed of loose packed ITO nanoparticles, shows excellent gas sensing properties toward NO2 and H2S, which gives the material potential applications in detecting hazardous gases.

Correctly predicting the site of O-glycosylation will greatly benefit the search and design of new specific and efficient GalNAc-transferase inhibitors. In this article, the site of O-glycosylation was studied using the correlation-based feature subset (CfsSubset) selection method combined with a wrapper method. Twenty-three important biochemical features were found based on a jackknife test from original data set containing 4779 features. By using the AdaBoost method with the twenty-three selected features, the prediction model yields an accuracy rate of 88.1% for the jackknife test and 87.5% for an independent set test, with increased accuracy over the original dataset by 8.5% and 10.42%, respectively. It is expected that our feature selection scheme can be referred to as a useful assistant technique for finding effective competitive inhibitors of GalNAc-transferase. An online predictor based on this research is available at http://chemdata.shu.edu.cn/gal_p/.► Fewer features (23 features) were selected from original data set (4779) features. ► Higher prediction accuracies obtain for jackknife test (88.1%) and independent set test (87.5%). ► Subsites P3, P1′ and P3′ are closely related to O-glycosylation. ► Secondary structure of amino acid residues shows high correlation to to O-glycosylation.

It is important to identify which proteins can interact with nucleic acids for the purpose of protein annotation, since interactions between nucleic acids and proteins involve in numerous cellular processes such as replication, transcription, splicing, and DNA repair. This research tries to identify proteins that can interact with DNA, RNA, and rRNA, respectively. mRMR (Minimum redundancy and maximum relevance), with its elegant mathematical formulation, has been applied widely in processing biological data and feature analysis since its introduction in 2005. mRMR plus incremental feature selection (IFS) is known to be very efficient in feature selection and analysis, and able to improve both effectiveness and efficiency of a prediction model. IFS is applied to decide how many features should be selected from feature list provided by mRMR. In the end, the selected features of mRMR and IFS are further refined by a conventional feature selection method—forward feature wrapper (FFW), by reordering the features. Each protein is coded by 132 features including amino acid compositions and physicochemical properties. After the feature selection, k-Nearest Neighbor algorithm, the adopted prediction model, is trained and tested. As a result, the optimized prediction accuracies for the DNA, RNA, and rRNA are 82.0, 83.4, and 92.3%, respectively. Furthermore, the most important features that contribute to the prediction are identified and analyzed biologically. The predictor, developed for this research, is available for public access at http://chemdata.shu.edu.cn/protein_na_mrmr/.

Dentritic neodymium hydroxycarbonate NdOHCO3 nanostructures have been successfully prepared using a facile hydrothermal approach, by the reaction of neodymium nitrate and ammonium hydrogen carbonate without the assistance of any surfactants or templates. The influences of the ratio of neodymium nitrate/ammonium hydrogen carbonate, and different reaction times and temperatures have been discussed. The as-obtained superstructures are characterized by several techniques, such as X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS), transmission electron microscopy (TEM), scanning electron microscopy (SEM), and fluorescence spectra measurements. The three-dimensional (3D) dentritic NdOHCO3 nanostructure is composed of one trunk with the diameter of about 50 nm and the length of about ca. 1−5 μm. And every branch is also a small dentritic nanostructure. The length of the secondary and tertiary sharp branches is about 300 nm to 2 μm and 100−300 nm, respectively. The photoluminescence (PL) spectra of the product shows an intensive band centered at about 388 nm. The relative peak intensity decreases as the reaction temperature increased and reaction time decreased. The possible formation mechanism of the 3D dentritic NdOHCO3 superstructures is proposed and discussed.

Atomic properties and ionic conductivity data of perovskite-type oxides were collected from literatures and experiments. The relationship between the electrical conductivity and the atomic property was examined. The oxide ionic conductivities were predicted by using two semi-empirical approaches based on first-principles calculations and three machine learning methods, such as partial least squares (PLS), back propagation artificial neural network (BP-ANN), and support vector regression (SVR). It was found that P/L (the ratio of O–O charge population to the O–O band length) has a quadratic curving relationship with Lnσ (logarithm of oxide ion conductivity) in some undoped perovskite-type oxides. The results of machine learning indicate that the generalization ability of SVR is better than those of BP-ANN and PLS models for predicting Lnσ.

AdaBoost Learner is employed to investigate Structure–Activity Relationships of phenols based on molecular descriptors. In this paper, the performance of AdaBoost Learner is compared with support vector machine (SVM), artificial neural networks (ANNs) and K nearest neighbors (KNNs), which are the most common algorithms used for SARs analysis. AdaBoost Learner performed better than SVM, ANNs and KNNs in predicting the mechanism of toxicity of phenols based on molecular descriptors. It can be concluded that AdaBoost has a potential to improve the performance of SARs analysis. We believe that AdaBoost Learner will play an important and complementary role to the existing algorithms for the prediction of the mechanisms of toxicity based on SARs. We have developed an online web server for the prediction of ecotoxicity mechanisms of phenols, accessible at http://chemdata.shu.edu.cn/ecotoxity/.

Large-scale CeO2 spherical architectures composed of numerous nanoflakes have been controllably prepared through a simple hydrothermal reaction without any template. The products were characterized with X-ray diffraction, nitrogen adsorption−desorption experiments, transmission electron microscopy (TEM), and high-resolution transmission electron microscopy (HRTEM). It was found that the CeO2 architecture ca. 100−230 nm in diameter was made up of many nanoflakes with a BET surface of 24 m2/g. The possible mechanism for the nanostructures formation was discussed. The catalytic performance of CeO2 nanospheres and the direct-depositing CeO2 nanoparticles in CO oxidation were also tested, and the catalytic results were compared and explained by analyzing the exposed planes of the two.

The knowledge of whether one enzyme can interact with a small molecule is essential for understanding the molecular and cellular functions of organisms. In this paper, we introduce a classifier to predict the small molecule– enzyme interaction, i.e., whether they can interact with each other. Small molecules are represented by their chemical functional groups, and enzymes are represented by their biochemical and physicochemical properties, resulting in a total of 160 features. These features are input into the AdaBoost classifier, which is known to have good generalization ability to predict interaction. As a result, the overall prediction accuracy, tested by tenfold cross-validation and independent sets, is 81.76% and 83.35%, respectively, suggesting that this strategy is effective. In this research, we typically choose interactions between small molecules and enzymes involved in metabolism to ultimately improve further understanding of metabolic pathways. An online predictor developed by this research is available at http://chemdata.shu.edu.cn/small_m.

Hierarchical cantaloupe-like and hollow microspherical AlOOH superstructures were successfully synthesized on a large scale via a one-step hydrothermal route. The as-obtained superstructures were characterized by several techniques, such as X-ray diffraction (XRD), transmission electron microscopy (TEM), scanning electron microscopy (SEM), and nitrogen adsorption/desorption measurement. The as-obtained superstructures, consisting of closely packed nanorods in an ordered fashion, have an average horizontal axis of ca. 2.5 µm and a longitudinal axis of ca. 1.5 µm. The as-obtained cantaloupe-like AlOOH superstructures have Brunauer–Emmett–Teller (BET) surface area of about 55.5 m2/g. The possible formation mechanism of the cantaloupe-like AlOOH superstructures is proposed and discussed.