Hierarchical learning architecture with automatic feature selection for multiclass protein fold classification

The structure classification of proteins plays a very important role in bioinformatics, since the relationships and characteristics among those known proteins can be exploited to predict the structure of new proteins. The success of a classification system depends heavily on two things: the tools being used and the features considered. For the bioinformatics applications, the role of appropriate features has not been paid adequate importance. In this investigation we use three novel ideas for multiclass protein fold classification. First, we use the gating neural network, where each input node is associated with a gate. This network can select important features in an online manner when the learning goes on. At the beginning of the training, all gates are almost closed, i.e., no feature is allowed to enter the network. Through the training, gates corresponding to good features are completely opened while gates corresponding to bad features are closed more tightly, and some gates may be partially open. The second novel idea is to use a hierarchical learning architecture (HLA). The classifier in the first level of HLA classifies the protein features into four major classes: all alpha, all beta, alpha + beta, and alpha/beta. And in the next level we have another set of classifiers, which further classifies the protein features into 27 folds. The third novel idea is to induce the indirect coding features from the amino-acid composition sequence of proteins based on the N-gram concept. This provides us with more representative and discriminative new local features of protein sequences for multiclass protein fold classification. The proposed HLA with new indirect coding features increases the protein fold classification accuracy by about 12%. Moreover, the gating neural network is found to reduce the number of features drastically. Using only half of the original features selected by the gating neural network can reach comparable test accuracy as that using all the original features. The gating mechanism also helps us to get a better insight into the folding process of proteins. For example, tracking the evolution of different gates we can find which characteristics (features) of the data are more important for the folding process. And, of course, it also reduces the computation time.     

Improved field emission from multiwall carbon nanotubes with nano-size defects produced by ultra-low energy ion bombardment

Multiwalled carbon nanotubes were irradiated with ultra-low energy (few eV) nitrogen and hydrogen ions using a microwave discharge. These ultra-low energy plasma-ions remain confined to the nanotube walls, transferring their maximum energy to the carbon atoms, and produce extraordinary structural changes to the carbon nanotube pillars as well as within the carbon nanotubes. Conical shaped emitters and nanotube structures with nano-defects are produced that exhibit remarkable field emission with ultra-low turn-on electric field (∼0.16 V/μm) and a >300-fold increase in the maximum emission current density compared to non-irradiated nanotubes. Doping of nitrogen is also identified due to such irradiation processes.     

Chemical modification of metallocene‐based polyethylene–octene elastomer through solution grafting of acrylic acid and its effect on the physico‐mechanical properties

In the present work, the metallocene‐based polyethylene–octene elastomer (POE) was chemically modified by solution grafting of acrylic acid in presence of benzoyl peroxide. The relative proportions of graft and gel formation were optimized through %weight gain, Fourier Transform infrared spectroscopy, elemental analysis and proton nuclear magnetic resonance spectroscopy. The gel formation in the POE matrix was found to be the prime competitor. The effect of grafting at its maximum level on various physicomechanical properties was thoroughly investigated, using X‐ray diffraction analysis, differential scanning calorimetry, mechanical, dynamic mechanical, and thermogravimetric analysis. © 2007 Wiley Periodicals, Inc. J Appl Polym Sci 2007     

Design and test of a Pareto optimal flat pad aerostatic bearing

This paper discusses the design and assessment of a Pareto optimal aerostatic flat pad bearing. The design of such bearings is a multi-variable, multi-criteria problem. The criteria used were load capacity, flow rate and stiffness. The design technique does not rely on weighting the criteria at the outset but allows a final design to be selected from the Pareto set. The final design was assessed using an automated experimental rig to obtain load/deflection data, mass flow rates, and film pressure profiles for a range of gap heights. The experimental performance of the Pareto optimal bearing is shown to compare well with the theoretical prediction at lower gap heights.     

Novel synthesis of ultra‐fine Sb2 O3 nanocubes using plant extract

In this study, the synthesis of ultra‐fine grade antimony trioxide (Sb₂ O₃) using plant extract for the first time is reported. Antimony chloride was used as a starting material and Dioscorea alata tuber extract was used as a reducing and capping agent. The synthesised nanoparticles were characterised by X‐ray diffraction (XRD), field emission scanning electron microscopy (FE‐SEM), dynamic light scattering (DLS), and Fourier transform infrared spectroscopy. XRD analysis indicates the formation of pure Sb₂ O₃ nanoparticles. The result from FE‐SEM and DLS showed that the particles have a cube‐like morphology and have an average size of 346.4 nm which falls within the range of ultra‐fine grade Sb₂ O₃.Inspec keywords: field emission electron microscopy, scanning electron microscopy, X‐ray diffraction, particle size, nanofabrication, light scattering, transmission electron microscopy, ultraviolet spectra, nanoparticles, antimony compounds, Fourier transform infrared spectraOther keywords: field emission scanning electron microscopy, FE‐SEM, dynamic light scattering, DLS, XRD analysis, antimony chloride, starting material, reducing agent, ultrafine grade antimony trioxide, plant extract, dioscorea alata tuber extract, capping agent, X‐ray diffraction, pure antimony trioxide nanoparticles, cube‐like morphology, Sb₂ O₃      

The differential induction machine: Theory and performance

This paper presents the theory and performance of a differential induction machine, which is a special type of induction machine having two shafts projected from the two ends of a single stator. Application of a differential load on the two shafts cause them to run at different speed as a motor, which permits true differential movement and thus can meet the requirements of a differential drive in an electric vehicle. The machine is also capable of regeneration in the differential mode. This paper presents the construction of the above machine and performance of the same based on experimental results from a laboratory prototype. The equivalent circuit of the motor has been presented and verified experimentally.     

Graphene-Reinforced Aluminum Matrix Composites: A Review of Synthesis Methods and Properties

Graphene-reinforced aluminum (Gr-Al) matrix nanocomposites (NCs) have attracted strong interest from both research and industry in high-performance weight-sensitive applications. Due to the vastly different bonding characteristics of the Al matrix (metallic) and graphene (in-plane covalent?+?inter-plane van der Waals), the graphene phase has a general tendency to agglomerate and phase separate in the metal matrix, which is detrimental for the mechanical and chemical properties of the composite. Thus, synthesis of Gr-Al NCs is extremely challenging. This review summarizes the different methods available to synthesize Gr-Al NCs and the resulting properties achieved in these NCs. Understanding the effect of processing parameters on the realized properties opens up the possibility of tailoring the synthesis methods to achieve the desired properties for a given application.     

Non-monotonic salt concentration dependence of inverted electrokinetic flow

Modeling of electrokinetic flows is crucial to understand numerous phenomena associated with electrochemistry, biophysics, and colloidal science. Here, we incorporate the modified Gaussian renormalized fluctuation theory into transport equations for electrolyte solutions to study the ion-correlation-induced inversion of electrokinetic flows, also known as charge inversion. We are able to capture the non-monotonic dependence of inverted streaming current and reversed electrophoretic mobility on salt concentration. By analyzing the double-layer structure, we elucidate that this non-monotonicity is a consequence of the competition between spatially varying ion correlations and the translational entropy of the ions. We find that for practical values of surface charge densities, the excluded volume effect does not play any significant role. In a significant improvement over existing theories, our theoretical predictions are in quantitative agreement with experimental measurements for charge inversion in trivalent salts.     

Bias caused by self-reporting distraction and its impact on crash estimates

Over the last decade, driver distractions, such as cell phone use and texting, have become a significant contributor to roadway crashes. Some states now have legislation that severely restricts or bans driver activities deemed distracting. However, many policies and engineered countermeasures are based on self-reported crash data. This raises the issue of potential bias and when not controlled for in analysis supporting policy decisions, can lead to poor allocation of public resources. This study explores the impact of self-reporting driver distraction on the likelihood estimates of the injury severity category of vehicle crashes. Using a two-step correction technique, the presence of bias is tested, when present corrected, and its impact is interpreted. The findings show that self-reporting bias is present in the national database, a database often used to help evaluate policy and engineering options, self-reporting bias understates the true effect of driver distraction on injury severity, and it is not uniform across injury categories. As a result, the forecast of potential savings of countermeasure policies or in-vehicle devices will be distorted leading to inefficient allocation of public resources.