Two cathelicidin genes are present in both rainbow trout (Oncorhynchus mykiss) and Atlantic salmon (Salmo salar)

Further to the previous finding of the rainbow trout rtCATH_1 gene, this paper describes three more cathelicidin genes found in salmonids: two in Atlantic salmon, named asCATH_1 and asCATH_2, and one in rainbow trout, named rtCATH_2. All the three new salmonid cathelicidin genes share the common characteristics of mammalian cathelicidin genes, such as consisting of four exons and possessing a highly conserved preproregion and four invariant cysteines clustered in the C-terminal region of the cathelin-like domain. The asCATH_1 gene is homologous to the rainbow trout rtCATH_1 gene, in that it possesses three repeat motifs of TGGGGGTGGC in exon IV and two cysteine residues in the predicted mature peptide, while the asCATH_2 gene and rtCATH_2 gene are homologues of each other, with 96% nucleotide identity. Salmonid cathelicidins possess the same elastase-sensitive residue, threonine, as hagfish cathelicidins and the rabbit CAP18 molecule. The cleavage site of the four salmonid cathelicidins is within a conserved amino acid motif of QKIRTRR, which is at the beginning of the sequence encoded by exon W. Two 36-residue peptides corresponding to the core part of rtCATH_1 and rtCATH_2 were chemically synthesized and shown to exhibit potent antimicrobial activity. rtCATH_2 was expressed constitutively in gill, head kidney, intestine, skin and spleen, while the expression of rtCATH_1 was inducible in gill, head kidney, and spleen after bacterial challenge. Four cathelicidin genes have now been characterized in salmonids and two were identified in hagfish, confirming that cathelicidin genes evolved early and are likely present in all vertebrates.

Single-electron tunneling depressing synapse for cellular neural networks

In this paper, a cellular neural network with depressing synapses for contrast-invariant pattern classification and synchrony detection is presented, starting from the impulse model of the single-electron tunneling junction. The results of the impulse model and the network are simulated using simulation program with integrated circuit emphasis (SPICE). It is demonstrated that depressing synapses should be an important candidate of robust systems since they exhibit a rapid depression of excitatory postsynaptic potentials for successive presynaptic spikes.

Electronic structure and exciton states in the freestanding ZnO nanorods

The electronic structure and exciton states of cylindrical ZnO nanorods with radius from 2 to 6 nm are investigated based on the framework of the effective-mass theory. Using the adiabatic approximation, the exciton binding energies taking account of the dielectric mismatch are solved exactly when the total angular momentum of the exciton states L = 0 and L = +/- 1. We find that the exciton binding energies can be enhanced greatly by the dielectric mismatch and the calculated results are almost consistent with the experimental data. Meanwhile, we obtain the optical transition rule when the small spin-obit splitting Delta(so) of ZnO is neglected. Furthermore, the radiative lifetime and linear optical susceptibilities chi(w) of the exciton states are calculated theoretically. The theoretical results are consistent with the experimental data very well. (C) 2009 American Institute of Physics. [DOI 10.1063/1.3125456]

Aharonov-Bohm oscillation and chirality effect in optical activity of single-wall carbon nanotubes

We study the Aharonov-Bohm effect in the optical phenomena of single-wall carbon nanotubes (SWCN) and also their chirality dependence. Especially, we consider the natural optical activity as a proper observable and derive its general expression based on a comprehensive symmetry analysis, which reveals the interplay between the enclosed magnetic flux and the tubule chirality for arbitrary chiral SWCN. A quantitative result for this optical property is given by a gauge invariant tight-binding approximation calculation to stimulate experimental measurements.

Electronic transport through self-assembled quantum disks

In this paper we study a single electron tunneling through a vertically stacked self-assembled quantum disks structure using a transfer matrix technique in the framework of effective mass approximation. In the disks, the electron is confined both laterally and vertically; we separate the motion in the vertical and lateral directions within the adiabatic approximation and treat the energy levels of the latter as an effective confining potential. The influence of a constant applied electric field is taken into account using an exact Airy-function formalism and the current density is calculated at zero temperature. By increasing the widths of the barriers, we find the peaks of the current density shift toward lower voltage region; meanwhile, they can become even sharper. (C) 2004 Elsevier Ltd. All rights reserved.

Digits speech recognition based on geometrical learning

We investigate the use of independent component analysis (ICA) for speech feature extraction in digits speech recognition systems.We observe that this may be true for a recognition tasks based on geometrical learning with little training data. In contrast to image processing, phase information is not essential for digits speech recognition. We therefore propose a new scheme that shows how the phase sensitivity can be removed by using an analytical description of the ICA-adapted basis functions via the Hilbert transform. Furthermore, since the basis functions are not shift invariant, we extend the method to include a frequency-based ICA stage that removes redundant time shift information. The digits speech recognition results show promising accuracy, Experiments show method based on ICA and geometrical learning outperforms HMM in different number of train samples.

Level-oscillation-induced pump effect in a quantum dot with asymmetric constrictions

We have investigated the pump effect induced by the level oscillation in a quantum dot with asymmetric constrictions. The curve of pumped current versus the frequency of level oscillation undulates at zero temperature. The oscillation of the pumped current can be smeared by increasing the temperature and the coupling strength between the quantum dot and the leads. Either the temperature increase or the coupling strength enhancement can lead to a positive or negative effect on the pumped current, depending on the parameters of the quantum dot system. A larger level-oscillation magnitude results in a larger pumped current, especially in the low-frequency case. An analytical expression of the pumped current is obtained in the regime far from adiabatic. A convenient physical picture based on our analytic result is proposed, with which we can explain all the features of the pumped current curves.

Excitonic energy of vertically stacked self-assmbled InAs quantum dots

We have studied the exciton states in vertically stacked self-assembled quantum disks within the effective mass approximation. The energy spectrum of the electron and hole is calculated using the transfer matrix formalism in the adiabatic approximation. The Coulomb interaction between the electron and the hole is treated accurately by the direct diagonalization of the Hamiltonian matrix. The effect of the vertical alignment of the disks on the ground energy of heavy- and light-hole exciton is presented and discussed. The binding energy is discussed in terms of the probability of the ground wave function. The ground energy of heavy- and light-hole excitons as a function of the magnetic field is presented and the effect of the disk size (the radius of disks) on the exciton energy is discussed.

Measurement of small-signal and large-signal responses of packaged laser modules at high temperature

In this paper, the pulsed injection method is extended to measure the chip temperature of various packaged laser modules, such as the DFB laser modules, the FP laser modules, and the EML laser modules. An optimal injection condition is obtained by investigating the dependence of the lasing wavelength on the width and period of the injection pulse in a relatively wide temperature range. The small-signal frequency responses and large-signal performances of packaged laser modules at different chip temperature are measured. The adiabatic small-signal modulation characteristics of packaged LD are first extracted. In the large-signal measurement, the effects of chip temperature, bias current and driving signal on the performances of the laser modules are discussed. It has been found that the large-signal performances of the EML modules depend on the different red-shift speeds of the DFB and EAM sections as chip temperature varying, and the optimal characteristics may be achieved at higher temperature.

Continuous speech recognition based on ICA and geometrical learning

We investigate the use of independent component analysis (ICA) for speech feature extraction in digits speech recognition systems. We observe that this may be true for recognition tasks based on Geometrical Learning with little training data. In contrast to image processing, phase information is not essential for digits speech recognition. We therefore propose a new scheme that shows how the phase sensitivity can be removed by using an analytical description of the ICA-adapted basis functions. Furthermore, since the basis functions are not shift invariant, we extend the method to include a frequency-based ICA stage that removes redundant time shift information. The digits speech recognition results show promising accuracy. Experiments show that the method based on ICA and Geometrical Learning outperforms HMM in a different number of training samples.

Theory of ultrafast optical manipulation of electron spins in quantum wells

Based on a multiparticle-state stimulated Raman adiabatic passage approach, a comprehensive theoretical study of the ultrafast optical manipulation of electron spins in quantum wells is presented. In addition to corroborating experimental findings [Gupta , Science 292, 2458 (2001)], we improve the expression for the optical-pulse-induced effective magnetic field, in comparison with the one obtained via the conventional single-particle ac Stark shift. Further study of the effect of hole-spin relaxation reveals that, while the coherent optical manipulation of electron spin in undoped quantum wells would deteriorate in the presence of relatively fast hole-spin relaxation, the coherent control in doped systems can be quite robust against decoherence. The implications of the present results on quantum dots will also be discussed. (c) 2005 American Institute of Physics.

Electronic structure of coupled vertically stacked self-assembled InAs quantum disks in a vertical electric field

In the framework of the effective-mass and adiabatic approximations, by setting the effective-mass of electron in the quantum disks (QDs) different from that in the potential barrier material, we make some improvements in the calculation of the electronic energy levels of vertically stacked self-assembled InAs QD. Comparing with the results when an empirical value was adopted as the effective-mass of electron of the system, we can see that the higher levels become heightened. Furthermore, the Stark shifts of the system of different methods are compared. The Stark shifts of holes are also studied. The vertical electric field changes the splitting between the symmetric level and the antisymmetric one for the same angular momentum. (C) 2003 Elsevier Ltd. All rights reserved.

Nonadiabatic geometric quantum computation with trapped ions

We propose a nonadiabatic scheme for geometric quantum computation with trapped ions. By making use of the Aharonov-Anandan phase, the proposed scheme not only preserves the globally geometric nature in quantum computation, but also provides the advantage of nonadiabaticity that overcomes the problem of slow evolution in the existing adiabatic schemes. Moreover, the present scheme requires only two atomic levels in each ion, making it an appealing candidate for quantum computation.

Continuous speech recognition based on ICA and geometrical learning

We investigate the use of independent component analysis (ICA) for speech feature extraction in digits speech recognition systems. We observe that this may be true for recognition tasks based on Geometrical Learning with little training data. In contrast to image processing, phase information is not essential for digits speech recognition. We therefore propose a new scheme that shows how the phase sensitivity can be removed by using an analytical description of the ICA-adapted basis functions. Furthermore, since the basis functions are not shift invariant, we extend the method to include a frequency-based ICA stage that removes redundant time shift information. The digits speech recognition results show promising accuracy. Experiments show that the method based on ICA and Geometrical Learning outperforms HMM in a different number of training samples.