Search

Multi-label learning tackles the problem in which each instance is composed of a single sample and associated with multiple labels simultaneously. In the past decades, many algorithms have been proposed for this emerging machine learning paradigm. However, these methods often focus on designing new classification strategies. If the original input data with a high dimensionality, it will suffer from the curse of dimensionality and get highly biased estimates. To address the issue, this paper proposes a novel feature construction approach named Structured Feature (SF), which can achieve a discriminant representation for multi-label data by exploiting the local and global geometric structures. These structures are characterized by the reconstruction coefficients of data in manifold and sparse subspaces. Since reconstruction coefficients reflect the data similarity relationships, and the similar data usually has the common class labels, SF incorporates the similarity relationships and supervised information into the feature representation. The SF contains supervised information, thus, even though classifier design ignores the correlations and interdependences between labels, it is able to obtain satisfactory classification performance.SF combines manifold and sparse structures in the feature representation for multi-label learning. Extensive experiments validate that SF can achieve comparable to or even better results than other state-of-the-art algorithms.

Therapeutic antibodies are a major class of pharmaceutical drugs used to treat a wide variety of diseases. They have several advantages including the high specificity and binding affinity to their molecular targets, and generally low immunotoxicity and mild adverse effects. The characterization of therapeutic antibodies is crucial to ensure drug efficacy and safety. Charge variant analysis can be used to examine the charge variant forms of therapeutic antibodies, which may reflect modifications that impact the drug quality. Native capillary electrophoresis-mass spectrometry (nCE-MS) analysis by an integrated ZipChip CE-MS system is an alternative and complementary method to cation-exchange chromatography and imaged capillary isoelectric focusing to support the characterization of charge variants. In this study, we performed nCE-MS analysis to evaluate the charge variants and impurities in therapeutic antibodies including immunoglobin G (IgG) monoclonal antibodies (mAbs), bispecific antibodies (bsAbs), and alternative formats such as therapeutic antibodies with addition or removal of antigen-binding domain. With the ZipChip CE-MS system, high-resolution charge variant separation was achieved for different types of therapeutic antibodies. Moreover, ZipChip nCE-MS analysis enabled high-sensitivity detection and identification of species with low abundance, including proteolytic cleavage and fragmentation in mAb, monospecific mAb impurities in bsAb, and O-glycosylation in alternative formats to support biopharmaceutical development and investigations.

On tokamaks, there are many diagnostics, which need real-time data acquisition and processing to provide useful information for plasma control. Some of the diagnostics required fast processing of multiple very high sampling rate signals. It is often difficult to achieve even with modern multi-core CPUs or GPUs. Moving large amount of data from the digitizers to the system ram will hurt the deterministic performance significantly. FPGA data pre-processing is a better choice when the data large and deterministic performance is required. When the task is too heavy for a single FPGA to implement, transfer large data between FPGAs is also required. On the polarimeter-interferometer diagnostics on J-TEXT, in order to calculate the density profile of the plasma in real-time, multiple chords of line-integrated density must be calculated first. The data acquisition system for the polarimeter-interferometer is equipped with 5 FlexRIO FPGAs, only one is acquiring the reference signal. Density calculation for every probing chord needs the reference signal. So this signal needs to be transferred to other FPGAs with strict real-time requirement. Mismatching a single sample, the result would be completely incorrect. NI P2P streaming allow the FPGA to transfer data without burdening the process. In the paper, the determinism characteristic and other performance of the P2P streaming is tested, and a mechanism to align local digitized samples with P2P streamed samples is presented. These results can be applied to other real-time data processing system that requires multiple FPGAs.

Liquid chromatography-mass spectrometry (LC-MS) is in wide use for compound identification and quantification in complex matrices. While advances in mass spectrometry and the incorporation of new acquisition methods have resulted in greatly improved detection, there is an ongoing need to expand the limits of highly sensitive and confident identification of low abundance species in complex samples. The data acquisition method known as "BoxCar" was originally designed to achieve in-depth proteome profiling on an Orbitrap mass analyzer by decomposing ions into segments with narrow m/z windows. Using this method, selected segments are packaged in C-trap and all ions are then sent to Orbitrap for detection. In this study, we developed a flexible BoxCar acquisition method by placing more segments in the low m/z range for small molecule profiling. This new MS1 acquisition method was successfully integrated with iterative data dependent MS/MS acquisition by generating an inclusion list of ions detected in the flexible BoxCar to trigger the fragmentation of parent ions. The developed acquisition method was applied to the analysis of cell culture media, which plays a key role in antibody production. This challenging goal is of critical importance, as none of the currently available methods provide a comprehensive understanding of how individual components, metabolites, and impurities associated with the cell culture process might influence recombinant antibody production. Even when present at relatively low abundance, some components or impurities in the cell culture medium could have a profound impact on the quality and titer of the antibodies produced. The complex soy hydrolysate cell culture medium used in antibody generation has not been fully characterized. Using the developed flexible BoxCar acquisition method, we achieved 90 % higher sensitivity in experiments designed to detect spiked chemical substances at low abundance at the MS1 level compared to the full scan method. Iterative data-dependent acquisition (DDA) based on the targeted inclusion list generated much higher quality MS2 spectra and facilitated confident identification of low-abundance compounds. Our method achieved a 50 % increase in MS2 coverage of compounds present at low concentrations compared to conventional DDA methods. The results of our study demonstrate that this data acquisition workflow can be easily operated on Orbitrap mass spectrometers and used as a highly effective approach to improve sensitivity and high-confidence small molecule profiling in soy hydrolysate-based cell culture medium and thus provides significant support for therapeutic monoclonal antibody production.

We present the study on multiplexed beams automatic alignment of high efficiency excimer laser. He-Cd laser with the wavelength of 325 nm is selected as the automatic alignment laser at first. And then, the crosshair arrays close to lens array are proposed and designed as the references of near-field and far-field, whereafter array beams are imaged and processed by specific fluorescence imaging system and region segmentation separately. Experiments are carried out in pre-amplifier II with three beams and double passes in excimer laser MOPA system. Results indicate that accuracy of the automatic alignment beam is 0.54% of the diameter of the windows. Meanwhile, the whole process of automatic alignment just takes 40 s, which ensures intelligent and high effective integration of automatic alignment system.

Broad and potent human monoclonal neutralizing antibodies have considerable potential in the prevention and treatment of acquired immunodeficiency syndrome (AIDS). To identify the key amino acid recognition site contacted with neutralizing antibody Y498, peptides were panned from the PhD-12 peptide library and predicted using online software. Then, four key amino acid sites, G367, D368, E370, and V372 located on the CD4 binding loop on gp120 of envelope of human immunodeficiency virus-1 (HIV-1), were found to determine the neutralization of antibody Y498. Residue E370 is in the deep part of the CD4 binding loop, which affects Y498-mediated neutralization. This form of recognition leads to a somewhat limiting neutralization spectrum of neutralizing antibody Y498, although it has some neutralization ability. Further study of the interactions between the neutralizing antibody Y498 and its epitope on the surface of the virus may facilitate vaccine development and so prevent new AIDS cases.

It is generally accepted that increasing the object distance can decrease the radiation influence of the self-luminous object in a schlieren system. This paper shows that a larger object distance results in a brighter image that exerts a greater influence on schlieren imaging because of overwhelming. To decrease the radiation influence, we propose to put the self-luminous object nearer to the second schlieren mirror, e.g. in the front focal plane, and spatially filter the radiation at the back focal plane of the second schlieren mirror by a diaphragm. An experiment was carried out, showing that the image irradiance contributed by self-luminescence can be reduced by several orders by this scheme.

Molecular dynamics (MD) simulations are performed to investigate the temperature relaxation between electrons and ions in a fully ionized, dense hydrogen plasma. We used HM (J. P. Hansen and I. R. McDonald) potential and introduced a truncated Coulomb interaction, which can avoid Coulomb catastrophe by choosing an appropriate cutting radius. The calculated results are compared with those from theoretical models (LS, GMS, BPS), whose applicability is also discussed. The effect of the interaction between ions and electrons on the temperature relaxation process is also investigated in the strong collision region. Finally, we discuss the effect of exchange interaction of electrons to the temperature relaxation.

Iron (Fe) plaque, defined as a film of poorly crystalline Fe oxides deposited on the surface of rice roots, potentially mediates paddy-soil N2O emissions. The aims of this study were to test if, and how, Fe plaque affects N2O production and reduction within a rice paddy soil. Rice seedlings were grown so that Fe plaque was either present or absent. With Fe plaque present, emissions of both N2O and N2 doubled, with the abundance of both Fe-redox bacteria and denitrifying functional genes elevated at the root-soil interface. Under hydroponic conditions, Fe plaque promoted N2O emissions in the presence of NO3− but not NH4+. In addition, chelating the Fe(II) eliminated the promoting effects of Fe plaque on N2O emission while Fe(II) addition to the Fe plaque-free roots increased N2O emission. These results demonstrate that Fe plaque promotes soil N2O emission and N loss predominately via Fe(II) oxidation-coupled denitrification. Our results indicate that Fe plaque is a hotspot for both N2O emission and N loss from paddy soils. Mitigation of N2O emission and N loss from paddy soils should consider methods to limit Fe plaque effects.

Using a fully implanted cortical electrical stimulation (CES) device with low-frequency burst impulse train, we investigated the effects of CES alone on behavioral recovery and surface density of dendritic structure in a rat model of middle cerebral artery occlusion (MCAO). After MCAO in rats, magnetic resonance imaging (MRI) was used to confirm cortex infarction and to identify a location for implantation of stimulating electrode over the peri-infarct cortex. The device was implanted on the 6th day after MCAO with CES then lasting for 16 days. The stimulation program consisted of two sessions lasting half an hour in the morning (0.65 mA, 0.13 microC/phase) and in the afternoon (0.5 mA, 0.1 microC/phase). The stimulator delivered biphasic charge balanced pulses (pulse width=200 micros) with various frequencies of 50 Hz, 20 Hz and 5 Hz in repeated 10-s blocks. Rats in the CES group (n=12) spend a much shorter time to regain preoperative levels of body weight (BW) than those in the no stimulation (NS) group (n=9). In behavioral tests, the rats in the CES group showed greater functional recovery compared to the NS group. Moreover, the functional improvement coincided with an increase in surface density of dendritic processes immunoreactive to microtubule-associated protein 2 (MAP2) in peri-infarct cortex. These results suggest the feasibility of the fully implanted CES device and the efficacy of the new stimulation protocol alone to improve functional outcome and cortical neuronal structural plasticity following focal cerebral ischemia in rats.

A major problem which hinders the applications of neural prostheses is the inconsistent performance caused by tissue responses during long-term implantation. The study investigated a new approach for improving the electrode-neural tissue interface. Hydrogel poly(vinyl alcohol)/poly(acrylic acid) interpenetrating polymer networks (PVA/PAA IPNs) were synthesized and tailored as coatings for poly(dimethylsiloxane) (PDMS) based neural electrodes with the aid of plasma pretreatment. Changes in the electrochemical impedance and maximum charge injection (Q(inj)) limits of the coated iridium oxide microelectrodes were negligible. Protein adsorption on PDMS was reduced by approximately 85% after coating. In the presence of nerve growth factor (NGF), neurite extension of rat pheochromocytoma (PC12) cells was clearly greater on PVA/PAA IPN films than on PDMS substrates. Furthermore, the tissue responses of PDMS implants coated with PVA/PAA IPN films were studied by 6-week implantation in the cortex of rats, which found that the glial fibrillary acidic protein (GFAP) immunoreactivity in animals (n=8) receiving coated implants was significantly lower (p

Several synthetic approaches to N- alkylated derivatives of 4-hydroxypyrrolidine-2-carboxylic acid and 4-hydroxypyrrolidine-2-acetic acid are described. The final compounds have been evaluated as potential inhibitors of the GABA transport proteins GAT-1 and GAT-3. The biological assays used were based on bovine material or porcine brain. As compared to the corresponding 4-unsubstituted compounds, the 4-hydroxypyrrolidine-2-carboxylic acid and 4-hydroxypyrrolidine-2-acetic acid derivatives showed a significant decrease in the inhibitory potency at both GAT-1 and GAT-3 with only four compounds having reasonable affinity to GAT-1 (IC 50 : 5.1, 6.6 and 9.4 μM) or GAT-3 (IC 50 : 19.9 μM), respectively. The biological data of the 4-hydroxypyrrolidine-2-acetic acid derivatives indicates that (2 S )-configuration at the C-2 position for potent inhibition of GAT-1 and (4 R )-configuration at the C-4 position for potent inhibition of GAT-3 may be crucial.