A two-stage gene selection scheme utilizing MRMR filter and GA wrapper

Gene expression data usually contain a large number of genes, but a small number of samples. Feature selection for gene expression data aims at finding a set of genes that best discriminates biological samples of different types. In this paper, we propose a two-stage selection algorithm for genomic data by combining MRMR (Minimum Redundancy–Maximum Relevance) and GA (Genetic Algorithm). In the first stage, MRMR is used to filter noisy and redundant genes in high-dimensional microarray data. In the second stage, the GA uses the classifier accuracy as a fitness function to select the highly discriminating genes. The proposed method is tested for tumor classification on five open datasets: NCI, Lymphoma, Lung, Leukemia and Colon using Support Vector Machine (SVM) and Naïve Bayes (NB) classifiers. The comparison of the MRMR-GA with MRMR filter and GA wrapper shows that our method is able to find the smallest gene subset that gives the most classification accuracy in leave-one-out cross-validation (LOOCV).     

Electronic,Magnetic, Electric and Magnetocaloric Properties of the Multiferroic Orthorhombic YMnO3

Journal of Superconductivity and Novel Magnetism - Multiferroic oxide materials have attracted much intention in recent years due to their application in different fields such as magnetic...     

Electrodeposition of nanostructured cuprous oxide on various substrates and their electrochemical and photoelectrochemical properties

Journal of Materials Science: Materials in Electronics - Nanostructured Cu2O films have been developed on various conductive substrates (FTO, ITO, and Mo) by a low-cost electrodeposition process...     

Borosilicate nanoparticles prepared by exothermic phase separation

Nanoparticles play an important role in chemical and biological sciences due to their ability to bind and concentrate many molecules on their surface. Polymers and silica are widely used to make nanoparticles, but efforts to make nanoparticles from borosilicate glass--which exhibits high tolerance to chemicals and solvents, combined with excellent mechanical and thermal stability--have proved unsuccessful. Here we show that borosilicate nanoparticles (100-500 nm in size) can be synthesized by simply mixing a silicon-boron binary oxide solution, prepared using non-aqueous organic solvents, with water. This induces a vigorous exothermic phase separation in which borosilicate nanoparticles burst out of a silica phase. In addition to potential applications in the life sciences, monodisperse borosilicate particles could also have applications in the production of photonic bandgap devices with high optical contrast, contrast agents for ultrasonic microscopy or chemical filtration membranes.     

Infrared spectroscopy analyses of air-CH4 or air-CO gas flows interacting with polycrystalline CeO2, La2O3 and Lu2O3 oxides

A comparative study of reactivity between air-CH4 or air-CO gas flows and CeO2, La2O3 and Lu2O3 rare earth oxides was performed using Fourier transform infrared spectroscopy analyses of CO2 gas resulted from the conversion of CH4 or CO gases. Polycrystalline samples of CeO2, La2O3 and Lu2O3 were first prepared by specific precipitation methods followed by low temperature calcination process. In the case of Lu2O3 oxide, a new specific route was proposed. Crystallite dimensions were determined by X-ray diffraction and transmission electron microscopy analyses. Morphologies were characterized using scanning electron microscopy. Specific surface areas were determined from Brunauer-Emmett-Teller (BET) technique. Using infrared spectroscopy analyses, the conversion rates of CH4 or CO into CO2 were determined from the evolutions of CO2 vibrational band intensities, as a function of time and temperature. It was clearly established that, despite its low specific surface, the Lu2O3 oxide presented the highest capacity of conversion of CH4 or CO into CO2.     

Segmentation and recognition system for unknown-length handwritten digit strings

The segmentation of handwritten digit strings into isolated digits remains a challenging task. The difficulty for recognizing handwritten digit strings is related to several factors such as sloping, overlapping, connecting and unknown length of the digit string. Hence, this paper aims to propose a segmentation and recognition system for unknown-length handwritten digit strings by combining several explicit segmentation methods depending on the configuration link between digits. Three segmentation methods are combined based on histogram of the vertical projection, the contour analysis and the sliding window Radon transform. A recognition and verification module based on support vector machine classifiers allows analyzing and deciding the rejection or acceptance each segmented digit image. Moreover, various submodules are included leading to enhance the robustness of the proposed system. Experimental results conducted on the benchmark dataset show that the proposed system is effective for segmenting handwritten digit strings without prior knowledge of their length comparatively to the state of the art.     

Embarked electrical network robust control based on singular perturbation model

This paper deals with an approach of modelling in view of control for embarked networks which can be described as strongly coupled multi-sources, multi-loads systems with nonlinear and badly known characteristics. This model has to be representative of the system behaviour and easy to handle for easy regulators synthesis. As a first step, each alternator is modelled and linearized around an operating point and then it is subdivided into two lower order systems according to the singular perturbation theory. RST regulators are designed for each subsystem and tested by means of a software test-bench which allows predicting network behaviour in both steady and transient states. Finally, the designed controllers are implanted on an experimental benchmark constituted by two alternators supplying loads in order to test the dynamic performances in realistic conditions.     

Electronic structure and thermoelectric properties of the thioantimonate FePb4Sb6S14

The title compound was synthesized and its thermoelectric properties investigated. Electronic structure calculations and electrical conductivity measurements show semiconducting behavior. The results of thermopower measurements are presented. The high thermopower motivated us to investigate the effects arising from chemical doping. Cobalt and tin doped variants were synthesized and their physical property measurements show improved electrical conductivity.     

Cooperative driving: an ant colony system for autonomous intersection management

Autonomous intersection management (AIM) is an innovative concept for directing vehicles through the intersections. AIM assumes that the vehicles negotiate the right-of-way. This assumption makes the problem of the intersection management significantly different from the usually studied ones such as the optimization of the cycle time, splits, and offsets. The main difficulty is to define a strategy that improves the traffic efficiency. Indeed, due to the fact that each vehicle is considered individually, AIM faces a combinatorial optimization problem that needs quick and efficient solutions for a real time application. This paper proposes a strategy that evacuates vehicles as soon as possible for each sequence of vehicle arrivals. The dynamic programming (DP) that gives the optimal solution is shown to be greedy. A combinatorial explosion is observed if the number of lanes rises. After evaluating the time complexity of the DP, the paper proposes an ant colony system (ACS) to solve the control problem for large number of vehicles and lanes. The complete investigation shows that the proposed ACS algorithm is robust and efficient. Experimental results obtained by the simulation of different traffic scenarios show that the AIM based on ACS outperforms the traditional traffic lights and other recent traffic control strategies.     

Chaotic mixing using source�Csink microfluidic flows in a PDMS chip

We present an active fixed-volume mixer based on the creation of multiple source–sink microfluidic flows in a polydimethylsiloxane (PDMS) chip without the need of external or internal pumps. To do so, four different pressure-controlled actuation chambers are arranged on top of the 5 μl volume of the mixing chamber. After the mixing volume is sealed/fixed by microfluidic valves made using ‘microplumbing technology’, a virtual source–sink pair is created by pressurizing one of the membranes and, at the same time, releasing the pressure of a neighboring one. The pressurized air deforms the thin membrane between the mixing and control chambers and creates microfluidic flows from the squeezed region (source) to the released region (sink) where the PDMS membrane is turned into the initial state. Several schemes of operation of virtual source–sink pairs are studied. In the optimized protocol, mixing is realized in just a sub-second time interval, thanks to the implementation of chaotic advection.