Semisupervised Dimensionality Reduction and Classification Through Virtual Label Regression

Semisupervised dimensionality reduction has been attracting much attention as it not only utilizes both labeled and unlabeled data simultaneously, but also works well in the situation of out-of-sample. This paper proposes an effective approach of semisupervised dimensionality reduction through label propagation and label regression. Different from previous efforts, the new approach propagates the label information from labeled to unlabeled data with a well-designed mechanism of random walks, in which outliers are effectively detected and the obtained virtual labels of unlabeled data can be well encoded in a weighted regression model. These virtual labels are thereafter regressed with a linear model to calculate the projection matrix for dimensionality reduction. By this means, when the manifold or the clustering assumption of data is satisfied, the labels of labeled data can be correctly propagated to the unlabeled data; and thus, the proposed approach utilizes the labeled and the unlabeled data more effectively than previous work. Experimental results are carried out upon several databases, and the advantage of the new approach is well demonstrated.

Partially Supervised Neighbor Embedding for Example-Based Image Super-Resolution

Neighbor embedding algorithm has been widely used in example-based super-resolution reconstruction from a single frame, which makes the assumption that neighbor patches embedded are contained in a single manifold. However, it is not always true for complicated texture structure. In this paper, we believe that textures may be contained in multiple manifolds, corresponding to classes. Under this assumption, we present a novel example-based image super-resolution reconstruction algorithm with clustering and supervised neighbor embedding (CSNE). First, a class predictor for low-resolution (LR) patches is learnt by an unsupervised Gaussian mixture model. Then by utilizing class label information of each patch, a supervised neighbor embedding is used to estimate high-resolution (HR) patches corresponding to LR patches. The experimental results show that the proposed method can achieve a better recovery of LR comparing with other simple schemes using neighbor embedding.

Discriminative Orthogonal Neighborhood-Preserving Projections for Classification

Orthogonal neighborhood-preserving projection (ONPP) is a recently developed orthogonal linear algorithm for overcoming the out-of-sample problem existing in the well-known manifold learning algorithm, i.e., locally linear embedding. It has been shown that ONPP is a strong analyzer of high-dimensional data. However, when applied to classification problems in a supervised setting, ONPP only focuses on the intraclass geometrical information while ignores the interaction of samples from different classes. To enhance the performance of ONPP in classification, a new algorithm termed discriminative ONPP (DONPP) is proposed in this paper. DONPP 1) takes into account both intraclass and interclass geometries; 2) considers the neighborhood information of interclass relationships; and 3) follows the orthogonality property of ONPP. Furthermore, DONPP is extended to the semisupervised case, i.e., semisupervised DONPP (SDONPP). This uses unlabeled samples to improve the classification accuracy of the original DONPP. Empirical studies demonstrate the effectiveness of both DONPP and SDONPP.

Supervised Gaussian Process Latent Variable Model for Dimensionality Reduction

The Gaussian process latent variable model (GP-LVM) has been identified to be an effective probabilistic approach for dimensionality reduction because it can obtain a low-dimensional manifold of a data set in an unsupervised fashion. Consequently, the GP-LVM is insufficient for supervised learning tasks (e. g., classification and regression) because it ignores the class label information for dimensionality reduction. In this paper, a supervised GP-LVM is developed for supervised learning tasks, and the maximum a posteriori algorithm is introduced to estimate positions of all samples in the latent variable space. We present experimental evidences suggesting that the supervised GP-LVM is able to use the class label information effectively, and thus, it outperforms the GP-LVM and the discriminative extension of the GP-LVM consistently. The comparison with some supervised classification methods, such as Gaussian process classification and support vector machines, is also given to illustrate the advantage of the proposed method.

Microarray based Raman spectroscopic detection with gold nanoparticle probes

microarray approach based on surface-enhanced Raman spectroscopic (SERS) was developed for detection of spotted peptide, peptide-protein or protein-antibody interaction. The procedure involves the attachment of peptide-capped gold nanoparticles followed by silver deposition for signal enhancement. The attachment of the gold nanoparticles is achieved by standard avidin-biotin chemistry. The well-known biomolecular recognition pairs, IgG/protein A and biotin/avidin, were used to demonstrate proof-of-concept of the SERS assay.

Gold nanoparticle-based surface enhanced Raman scattering spectroscopic assay for the detection of protein-protein interactions

In the present work, a sensitive spectroscopic assay based on surface-enhanced Raman spectroscopy (SERS) using gold nanoparticles as substrates was developed for the rapid detection protein-protein interactions. Detection is achieved by specific binding biotin-modification antibodies with protein-stabilized 30 nm gold nanoparticles, followed by the attachment of avidin-modification Raman-active dyes. As a proof-of-principle experiment, a well-known biomolecular recognition system, IgG with protein A, was chosen to establish this new spectroscopic assay. Highly selective recognition of IgG down to 1 ng/ml in solution has been demonstrated.

Ethylene Polymerization of the New Titanium Complexes Bearing a Phosphine Oxide-Bridged Bisphenolato Ligand

A series of novel titanium(IV) complexes combining a phosphine oxide-bridged bisphenolato ligand TiCl2{2,2'-O=P-R-3 (4-R-2-6-R-1-C6H2O)(2)}(THF) (6a: R-1 = tBu, R-2 - H, R-3 Ph; 6b: R-1 - Ph, R-2 = H, R-3 = Ph; 6c: R-1 = R-2 = tBu, R-3 = Ph; 6d: R-1 = R-2 cumyl, R-3 = Ph; 6e: R-1 = tBu, R-2 = H, R-3 = PhF5) were prepared by the reaction of corresponding bisphenolato ligands with TiCl4 in THF. X-ray analysis reveals that complex 6a adopts distorted octahedral geometry around the titanium center. These catalysts were performed for ethylene polymerization in the presence of modified methyaluminoxane (MMAO).

Studies on the Biotransformation of Arctigenin Using Electrospray Ionization Mass Spectrometry

To study the biotransformation of arctigenin, arctigenin was anaerobically incubated with Eubacterium sp. ARC-2 of human intestinal bacteria in vitro. Arctigenin formed a molecular ion [M-H](-) in negative ion mode. The arctigenin and its metabolites were investigated directly by the electrospray ionization tandem mass spectrometry ion trap and Fourier transform ion cyclotron resonance. Arctigenin was transformed to 4',4 ''-dihydroxylenterolactone by E sp. ARC-2 through 3 types of demethylation products.

Ethylene Polymerization by the New Chromium Catalysts Based on Amino-Pyrrolide Ligands

A series of amino-pyrrolide ligands (1-4a) and their derivatives aminothiophene ligand (5a), amino-indole ligand (6a) were prepared. Chromium catalysts, which were generated in situ by mixing the ligands with CrCl3(thf)(3) in toluene, were tested for ethylene polymerization. The preliminary screening results revealed that the tridentate amino-pyrrolide ligands containing soft pendant donor, 3a, 4a/CrCl3(thf)(3) systems displayed high catalytic activities towards ethylene polymerization in the presence of modified methyaluminoxane. The electronic and steric factors attached to the ligand backbone significantly affected both the catalyst activity and the polymer molecular weight. Complex 4b was obtained by the reaction of CrCl3(thf)(3) with one equivalent of the lithium salts of 4a, which was the most efficient ligand among the tested ones. The effect of polymerization parameters such as cocatalyst concentration, ethylene pressure, reaction temperature, and time on polymerization behavior were investigated in detail. The resulting polymer obtained by 4b display wax-like and possess linear structure, low molecular weight, and unimodal distribution.

Ethylene Polymerization and Ethylene/Hexene Copolymerization with Vanadium(III) Catalysts Bearing Heteroatom-Containing Salicylaldiminato Ligands

A series of novel vanadium(III) complexes hearing heteroatoill-containing group-substituted salicylaldiminato ligands [RN=CH(ArO)]VCl2(THF)(2) (Ar = C6H4, R = C3H2NS, 2a; C7H4NS, 2c; C7H5N2, 2d; Ar = C(6)H(2)tBu(2) (2,4), R = C3H2NS, 2b) have been synthesized and characterized. Structure of complex 2c was further confirmed by X-ray crystallographic analysis. The complexes were investigated as the catalysts for ethylene polymerization in the presence of Et2AlCl. Complexes 2a-d exhibited high catalytic activities (up to 22.8 kg polyethylene/mmolv h bar), and affording polymer with unimodal molecular weight distributions at 25-70 degrees C in the first 5-min polymerization, whereas produced bimodal molecular weight distribution polymers at 70 degrees C when polymerization time prolonged to 30 min. The catalyst structure plays an important role in controlling the molecular weight and molecular weight distribution of the resultant polymers produced in 30 min polymerization. In addition, ethylene/hexene copolymerizations with catalysts 2a-d were also explored in the presence of Et2AlCl, which leads to the high molecular weight and unimodal distributions copolymers with high comonomer incorporation.

Synthesis of Titanate-Based Nanotubes for One-Dimensionally Confined Electrical Properties

Structural tailoring for dimensionally confined electrical properties is fundamentally important for nanodevices and the relevant technologies. Titanate-based nanotubes were taken as a prototype one-dimensional material to study. First, Na0.96H1.04Ti3O7 center dot 3.42H(2)O nanotubes were prepared by a simple hydrothermal condition, which converted into Na0.036H1.964Ti3O7 center dot 3.52H(2)O nanotubes by a subsequent acidic rinsing. Systematic sample characterization using combined techniques of X-ray diffraction, field emission scanning electron microscopy, high resolution transmission electron microscopy, electron paramagnetic resonance, Fourier transform infrared spectroscopy, elemental analyses, and alternative current impedance indicated that both nanotubes possessed a scrolled trititanate-type structure with the (200) crystal face predominant on the tube surface. With increasing temperature, both nanotubes underwent a continuous dehydration process, which however imposed different impacts oil the structures and electrical properties, depending on the types of the nanotubes

Study on the lifetime of M-intermediate of bacteriorhodopsin film chemically modified with CeO2 nanoparticles

The characteristics of intermediates of bacteriorhodopsin (bR) can be verified by chemical modification of its surroundings. CeO2 nanoparticles, which were obtained using water-in-oil (W/O) microemulsion and calcined at various temperatures, were used as chemical additive for the modification of bR. X-ray diffraction (XRD) shows that the mean particle sizes for the samples calcined at 500 and 800 degrees C are approximately 10 and 30 nm, respectively. We prepared CeO2 nanoparticle modified poly(vinyl alcohol) (bR-PVA) films with an optical density of about 1.5 at the ground state. It is observed that the lifetime of the Wintermediate for the modified films is prolonged compared with that of the unmodified ones, and the lifetime increases with decreasing particle size. A probable mechanism, which is likely to involve effective molecular interactions between the CeO2 nanoparticles and the bR molecules, is discussed. The hydroxyl groups, which might arise from the interaction between the nanoparticles and the surrounding water molecules, help to lower the ability of the Schiff base of uptaking protons in the Wintermediate. The results indicate that controlling the interactions between biomolecules and various nanomaterials would enlarge the functionality and the range of the application of nanoparticles.

Interaction of echinomycin with guanine: electrochemistry and spectroscopy studies

The interaction of antitumor antibiotic, echinomycin (Echi) with guanine (Gua) was thoroughly investigated by adsorptive transfer stripping cyclic voltammetry, ultraviolet and visible adsorption spectra (UV/Vis) and Fourier-transform infrared spectroscopy (FTIR). Electrochemistry provided a simple tool for verifying the occurrence of interaction between Echi and Gua. Echi could be accumulated from the solution and give well-defined electrochemical signals in 0.1 M phosphate buffer solution (pH 7.0) only when Gua was present on the surface of the electrochemically pretreated glass carbon electrode (GCE), suggesting a strong binding of Echi to Gua. All the acquired spectral data showed that a new adduct between Echi and Gua was formed, and two pairs of adjacent intermolecular hydrogen bonds between the Ala backbone atoms in Echi and Gua (Ala-NH to Gua-N3 and Gua-NH2 to Ala-CO) played a dominating role in the interaction. Electrochemistry coupled with spectroscopy techniques could provide a relatively easy way to obtain useful insights into the molecular mechanism of drug-DNA interactions, which should be important in the development of new anticancer drugs with specific base recognition.

Direct electrochemical generation of conducting polymer micro-rings

Self-doped polyaniline (PANI) micro-rings have been successfully generated electrochemically. The polymer forming rings were about 100 nm wide, and the ring diameter is tunable from several to dozens of micrometres depending on deferent current densities. The morphology of such nanostructured polyaniline rings was investigated and further confirmed with field-emission scanning electron microscopy (FE-SEM). Furthermore, the film was characterized using UV/visible spectroscopy and cyclic voltammetry. The bubble template formation mechanism of the micro-rings was also proposed. Such nanostructured materials synthesized electrochemically open up a new approach to surface morphology control.