Implementation of a hybrid PCM encoding scheme

An improved encoding scheme is proposed wherein the quantizer step size incorporates both instantaneous and syllabic adaptation. The information needed to carry out this adaptation is extracted at the coder after the D/A conversion of the digital output, so that the coder and the decoder track each other. A 5-bit coder has been designed and built which has a dynamic range of 38 dB with a constant SNR of 30 dB, and the quality of its output signal is found to be comparable to that of other coding schemes

Eliciting and encoding expert knowledge on variable selection into classical or Bayesian species distribution models

The quality of species distribution models (SDMs) relies to a large degree on the quality of the input data, from bioclimatic indices to environmental and habitat descriptors (Austin, 2002). Recent reviews of SDM techniques, have sought to optimize predictive performance e.g. Elith et al., 2006. In general SDMs employ one of three approaches to variable selection. The simplest approach relies on the expert to select the variables, as in environmental niche models Nix, 1986 or a generalized linear model without variable selection (Miller and Franklin, 2002). A second approach explicitly incorporates variable selection into model fitting, which allows examination of particular combinations of variables. Examples include generalized linear or additive models with variable selection (Hastie et al. 2002); or classification trees with complexity or model based pruning (Breiman et al., 1984, Zeileis, 2008). A third approach uses model averaging, to summarize the overall contribution of a variable, without considering particular combinations. Examples include neural networks, boosted or bagged regression trees and Maximum Entropy as compared in Elith et al. 2006. Typically, users of SDMs will either consider a small number of variable sets, via the first approach, or else supply all of the candidate variables (often numbering more than a hundred) to the second or third approaches. Bayesian SDMs exist, with several methods for eliciting and encoding priors on model parameters (see review in Low Choy et al. 2010). However few methods have been published for informative variable selection; one example is Bayesian trees (O’Leary 2008). Here we report an elicitation protocol that helps makes explicit a priori expert judgements on the quality of candidate variables. This protocol can be flexibly applied to any of the three approaches to variable selection, described above, Bayesian or otherwise. We demonstrate how this information can be obtained then used to guide variable selection in classical or machine learning SDMs, or to define priors within Bayesian SDMs.

Multi-species sequence comparison reveals conservation of ghrelin gene-derived splice variants encoding a truncated ghrelin peptide

The peptide hormone ghrelin is a potent orexigen produced predominantly in the stomach. It has a number of other biological actions, including roles in appetite stimulation, energy balance, the stimulation of growth hormone release and the regulation of cell proliferation. Recently, several ghrelin gene splice variants have been described. Here, we attempted to identify conserved alternative splicing of the ghrelin gene by cross-species sequence comparisons. We identified a novel human exon 2-deleted variant and provide preliminary evidence that this splice variant and in1-ghrelin encode a C-terminally truncated form of the ghrelin peptide, termed minighrelin. These variants are expressed in humans and mice, demonstrating conservation of alternative splicing spanning 90 million years. Minighrelin appears to have similar actions to full-length ghrelin, as treatment with exogenous minighrelin peptide stimulates appetite and feeding in mice. Forced expression of the exon 2-deleted preproghrelin variant mirrors the effect of the canonical preproghrelin, stimulating cell proliferation and migration in the PC3 prostate cancer cell line. This is the first study to characterise an exon 2-deleted preproghrelin variant and to demonstrate sequence conservation of ghrelin gene-derived splice variants that encode a truncated ghrelin peptide. This adds further impetus for studies into the alternative splicing of the ghrelin gene and the function of novel ghrelin peptides in vertebrates.

Two-Dimensional Encoding Masks ror Hadamard Spectrometric Imager

Techniques to improve upon the design of two-dimensional encoding masks for multiplex Hadamard spectrometric imagers are described. The technique to generate masks based on completely orthonormal codes is described. In some cases, selfsupporting masks result which were not previously known. Transmission through the encoding masks (but not necessarily the signalto-noise ratio) can also be increased.

Conformal invariance and scalar-tensor theories

A new generalisation of Einstein’s theory is proposed which is invariant under conformal mappings. Two scalar fields are introduced in addition to the metric tensor field, so that two special choices of gauge are available for physical interpretation, the ‘Einstein gauge’ and the ‘atomic gauge’. The theory is not unique but contains two adjustable parameters ζ anda. Witha=1 the theory viewed from the atomic gauge is Brans-Dicke theory (ω=−3/2+ζ/4). Any other choice ofa leads to a creation-field theory. In particular the theory given by the choicea=−3 possesses a cosmological solution satisfying Dirac’s ‘large numbers’ hypothesis.

Unitary Processes with Independent Increments and Representations of Hilbert Tensor Algebras

The aim of this article is to characterize unitary increment process by a quantum stochastic integral representation on symmetric Fock space. Under certain assumptions we have proved its unitary equivalence to a Hudson-Parthasarathy flow.

Analysis of the fusA2 locus encoding EFG2 in Mycobacterium smegmatis

The translation elongation factor G (EFG) is encoded by the fusA gene.Several bacteria possess a second fusA-like locus,fusA2 which encodes EFG2. A comparison of EFG and EFG2 from various bacteria reveals that EFG2 preserves domain organization and maintains significant sequence homology with EFG, suggesting that EFG2 may function as an elongation factor. However, with the single exception of a recent study on Thermus thermophilus EFG2, this class of EFG-like factors has not been investigated. Here, we have characterized EFG2 (MSMEG_6535) from Mycobacterium smegmatis. Expression of EFG2 was detected in stationary phase cultures of M.smegmatis (Msm). Our in vitro studies show that while MsmEFG2 binds guanine nucleotides, it lacks the ribosome-dependent GTPase activity characteristic of EFGs. Furthermore,unlike MsmEFG (MSMEG_1400), MsmEFG2 failed to rescue an E. coli strain harboring a temperature-sensitive allele of EFG, for its growth at thenon-permissive temperature. Subsequent experiments showed that the fusA2 gene could be disrupted in M. smegmatis mc(2)155 with Kan(R)marker. The M. smegmatis fusA2::kan strain was viable and showed growth kinetics similar to that of the parent strain (wild-type for fusA2).However, in the growth competition assays, the disruption of fusA2 was found to confer a fitness disadvantage to M. smegmatis, raising the possibility that EFG2 is of some physiological relevance to mycobacteria.

R1STM: One-class support tensor machine with randomised kernel

Identifying unusual or anomalous patterns in an underlying dataset is an important but challenging task in many applications. The focus of the unsupervised anomaly detection literature has mostly been on vectorised data. However, many applications are more naturally described using higher-order tensor representations. Approaches that vectorise tensorial data can destroy the structural information encoded in the high-dimensional space, and lead to the problem of the curse of dimensionality. In this paper we present the first unsupervised tensorial anomaly detection method, along with a randomised version of our method. Our anomaly detection method, the One-class Support Tensor Machine (1STM), is a generalisation of conventional one-class Support Vector Machines to higher-order spaces. 1STM preserves the multiway structure of tensor data, while achieving significant improvement in accuracy and efficiency over conventional vectorised methods. We then leverage the theory of nonlinear random projections to propose the Randomised 1STM (R1STM). Our empirical analysis on several real and synthetic datasets shows that our R1STM algorithm delivers comparable or better accuracy to a state-of-the-art deep learning method and traditional kernelised approaches for anomaly detection, while being approximately 100 times faster in training and testing.

Differential transcription of multiple copies of a silk worm gene encoding tRNA

Ten different tRNAGly1 genes from the silk worm, Bombyx mori, have been cloned and characterized. These genes were transcribed in vitro in homologous nuclear extracts from the posterior silk gland (PSG) or nuclear extracts derived from the middle silk gland or ovarian tissues. Although the transcription levels were much higher in the PSG nuclear extracts, the transcriptional efficiency of the individual genes followed a similar pattern in all the extracts. Based on the levels of in vitro transcription, the ten tRNAGly1 genes could be divided into three groups, viz., those which were transcribed at very high levels (e.g., clone pR8), high to medium levels (e.g., pBmil, pBmpl, pBmhl, pBmtl) and low to barely detectable levels (e.g., pBmsl, pBmjl and pBmkl). The coding sequences of all these tRNA genes being identical, the differential transcription suggested that the flanking sequences modulate their transcriptional efficiency. The presence of positive and negative regulatory elements in the 5' flanking regions of these genes was confirmed by transcription competition experiments. A positive element was present in the immediate upstream A + T-rich sequences in all the genes, but no consensus sequences correlating to the transcriptional status could be generated. The presence of negative elements on the other hand was indicated only in some of the genes and therefore may have a role in the differential transcription of these tRNAGly genes in vivo.

Identification of Mxr1p-binding sites in the promoters of genes encoding dihydroxyacetone synthase and peroxin 8 of the methylotrophic yeast Pichia pastoris

Expression of genes involved in methanol metabolism of Pichia pastoris is regulated by Mxr1p, a zinc finger transcription factor. In this study, we studied the target gene specificity of Mxr1p by examining its ability to bind to promoters of genes encoding dihydroxyacetone synthase (DHAS) and peroxin 8 (PEX8), since methanol-inducible expression of these genes is abrogated in mxr1-null mutant strains of P. pastoris. Different regions of DHAS and PEX8 promoter were isolated from P. pastoris genomic DNA and their ability to bind to a recombinant Mxr1p protein containing the N-terminal 150 amino acids, including the zinc finger DNA-binding domain, was examined. These studies reveal that Mxr1p specifically binds to promoter regions containing multiple 5'-CYCC-3' sequences, although all DNA sequences containing the 5'-CYCC-3' motif do not qualify as Mxr1p-binding sites. Key DNA-binding determinants are present outside 5'-CYCC-3' motif and Mxr1p preferably binds to DNA sequences containing 5'-CYCCNY-3' than those containing 5'-CYCCNR-3' sequences. This study provides new insights into the molecular determinants of target gene specificity of Mxr1p, and the methodology described here can be used for mapping Mxr1p-binding sites in other methanol-inducible promoters of P. pastoris. Copyright (C) 2010 John Wiley & Sons, Ltd.

Top-spin analysis of new scalar and tensor interactions in e(+)e(-) collisions with beam polarization

We utilize top polarization in the process e(+)e(-) -> t (t) over bar at the International Linear Collider ( ILC) with transverse beam polarization to probe interactions of the scalar and tensor type beyond the standard model and to disentangle their individual contributions. Ninety percent confidence level limits on the interactions with realistic integrated luminosity are presented and are found to improve by an order of magnitude compared to the case when the spin of the top quark is not measured. Sensitivities of the order of a few times 10(-3) TeV-2 for real and imaginary parts of both scalar and tensor couplings at root s = 500 and 800 GeV with an integrated luminosity of 500 fb(-1) and completely polarized beams are shown to be possible. A powerful model-independent framework for inclusive measurements is employed to describe the spin-momentum correlations, and their C, P, and T properties are presented in a technical appendix.

Memory Bandwidth and Power Reduction Using Lossy Reference Frame Compression in Video Encoding

Large external memory bandwidth requirement leads to increased system power dissipation and cost in video coding application. Majority of the external memory traffic in video encoder is due to reference data accesses. We describe a lossy reference frame compression technique that can be used in video coding with minimal impact on quality while significantly reducing power and bandwidth requirement. The low cost transformless compression technique uses lossy reference for motion estimation to reduce memory traffic, and lossless reference for motion compensation (MC) to avoid drift. Thus, it is compatible with all existing video standards. We calculate the quantization error bound and show that by storing quantization error separately, bandwidth overhead due to MC can be reduced significantly. The technique meets key requirements specific to the video encode application. 24-39% reduction in peak bandwidth and 23-31% reduction in total average power consumption are observed for IBBP sequences.

Speedup Macroblock Mode Decision in H.264/SVC Encoding Using Cost-Sensitive Learning

In this paper we present a novel macroblock mode decision algorithm to speedup H.264/SVC Intra frame encoding. We replace the complex mode-decision calculations by a classifier which has been trained specifically to minimize the reduction in RD performance. This results in a significant speedup in encoding. The results show that machine learning has a great potential and can reduce the complexity substantially with negligible impact on quality. The results show that the proposed method reduces encoding time to about 70% in base layer and up to 50% in enhancement layer of the reference implementation with a negligible loss in quality.

Cloning and sequence of a cDNA encoding a novel hybrid prolme-rich protein associated with cytokinin-induced haustoria formation in Cuscuta reflexa

A complete cDNA encoding a novel hybrid Pro-rich protein (HyPRP) was identified by differentially screening 3x10(4) recombinant plaques of a Cuscuta reflexa cytokinin-induced haustorial cDNA library constructed in lambda gt10. The nucleotide (nt) sequence consists of: (i) a 424-bp 5'-non coding region having five start codons (ATGs) and three upstream open reading frames (uORFs); (ii) an ORF of 987 bp with coding potential for a 329-amino-acid (aa) protein of M(r), 35203 with a hydrophobic N-terminal region including a stretch of nine consecutive Phe followed by a Pro-rich sequence and a Cys-rich hydrophobic C terminus; and (iii) a 178-bp 3'-UTR (untranslated region). Comparison of the predicted aa sequence with the NBRF and SWISSPROT databases and with a recent report of an embryo-specific protein of maize [Jose-Estanyol et al., Plant Cell 4 (1992) 413-423] showed it to be similar to the class of HyPRPs encoded by genes preferentially expressed in young tomato fruits, maize embryos and in vitro-cultured carrot embryos. Northern analysis revealed an approx. 1.8-kb mRNA of this gene expressed in the subapical region of the C. reflexa vine which exhibited maximum sensitivity to cytokinin in haustorial induction.