Search

[EN] In this paper, we propose, describe, and test a modification of the K-SVD algorithm. Given a set of training data, the proposed algorithm computes an overcomplete dictionary by minimizing the ß-divergence () between the data and its representation as linear combinations of atoms of the dictionary, under strict sparsity restrictions. For the special case , the proposed algorithm minimizes the Frobenius norm and, therefore, for the proposed algorithm is equivalent to the original K-SVD algorithm. We describe the modifications needed and discuss the possible shortcomings of the new algorithm. The algorithm is tested with random matrices and with an example based on speech separation., This work has been partially supported by the EU together with the Spanish Government through TEC2015-67387-C4-1-R (MINECO/FEDER) and by Programa de FPU del Ministerio de Educacion, Cultura y Deporte FPU13/03828 (Spain).

A method for binaural rendering of sound scene recordings is proposed.Source signals and their direction of arrival is estimated using a microphone array.A low-rank NMF model for separation of sound sources is used.Speech intelligibility test with overlapping speech is conducted.The speech intelligibility by binaural processing is shown to increase over stereo. This paper proposes a method for binaural reconstruction of a sound scene captured with a portable-sized array consisting of several microphones. The proposed processing is separating the scene into a sum of small number of sources, and the spectrogram of each of them is in turn represented as a small number of latent components. The direction of arrival (DOA) of each source is estimated, which is followed by binaural rendering of each source at its estimated direction. For representing the sources, the proposed method uses low-rank complex-valued non-negative matrix factorization combined with DOA-based spatial covariance matrix model. The binaural reconstruction is achieved by applying the binaural cues (head-related transfer function) associated with the estimated source DOA to the separated source signals. The binaural rendering quality of the proposed method was evaluated using a speech intelligibility test. The test results indicated that the proposed binaural rendering was able to improve the intelligibility of speech over stereo recordings and separation by minimum variance distortionless response beamformer with the same binaural synthesis in a three-speaker scenario. An additional listening test evaluating the subjective quality of the rendered output indicates no added processing artifacts by the proposed method in comparison to unprocessed stereo recording.

We provide an overview of matrix and tensor factorization methods from a Bayesian perspective, giving emphasis on both the inference methods and modeling techniques. Factorization based models and their many extensions such as tensor factorizations have proved useful in a broad range of applications, supporting a practical and computationally tractable framework for modeling. Especially in audio processing, tensor models help in a unified manner the use of prior knowledge about signals, the data generation processes as well as available data from different modalities. After a general review of tensor models, we describe the general statistical framework, give examples of several audio applications and describe modeling strategies for key problems such as deconvolution, source separation, and transcription. (C) 2015 Elsevier Inc. All rights reserved.

Four out of six officially recognized dog allergens are members of the lipocalin protein family. So far, a three-dimensional structure has been determined for only one dog allergen, Can f 2, which is a lipocalin protein. We present here the crystal structure of a second lipocalin allergen from dog, a variant of Can f 4. Moreover, we have compared and analyzed the structures of these two weakly homologous (amino acid identity 21%) dog allergens. The size and the amino acid composition of the ligand-binding pocket indicate that Can f 4 is capable of binding only relatively small hydrophobic molecules which are different from those that Can f 2 is able to bind. The crystal structure of Can f 4 contained both monomeric and dimeric forms of the allergen, suggesting that Can f 4 is able to form transient (weak) dimers. The existence of transient dimers in solution was confirmed by use of native mass spectrometry. The dimeric structure of Can f 4 is formed when the ends of four β-strands are packed against the same strands from the second monomer. The residues in the interface are mainly hydrophobic and the formation of the dimer is similar to the major horse allergen Equ c 1. Interestingly, the crystal structure of dog Can f 2 has been reported to show a different type of dimer formation. The capability of these allergens to form dimers may be important for the development of immediate allergic reaction (mast cell activation) because oligomeric allergens can effectively present multivalent epitopes.

Speech recognition systems intended for everyday use must be able to cope with a large variety of noise types and levels, including highly non-stationary multi-source mixtures. This study applies spectral factorisation algorithms and long temporal context for separating speech and noise from mixed signals. To adapt the system to varying environments, noise models are acquired from the context, or learnt from the mixture itself without prior information. We also propose methods for reducing the size of the bases used for speech and noise modelling by 20-40 times for better practical applicability. We evaluate the performance of the methods both as a standalone classifier and as a signal-enhancing front-end for external recognisers. For the CHiME noisy speech corpus containing non-stationary multi-source household noises at signal-to-noise ratios ranging from +9 to -6dB, we report average keyword recognition rates up to 87.8% using a single-stream sparse classification algorithm.

Background Peptide immunotherapy is a promising alternative for treating allergic diseases. One way to enhance the efficacy of peptide immunotherapy is to use altered peptide ligands (APLs) that contain amino acid substitutions compared with the natural peptide. Objective To evaluate the potential of an APL of the immunodominant epitope of lipocalin allergen Bos d 2 for peptide immunotherapy. Methods Peripheral blood CD4 + T-cell responses of 8 HLA-DR4–positive subjects to the natural ligand of Bos d 2 (p127-142) or to an APL (pN135D) were analyzed by MHC class II tetramer staining after in vitro expansion with the peptides. Long-term T-cell lines (TCLs) were induced with the peptides, and the cytokine production, cross-reactivity, and T-cell receptor Vβ subtype expression of the TCLs were analyzed. Results CD4 + T cells specific for both p127-142 and pN135D were readily detected in peripheral blood after a single in vitro stimulation. Whereas the TCLs induced with p127-142 were T H 2/T H 0-deviated, those induced with pN135D were T H 1/T H 0-deviated and highly cross-reactive with p127-142. Moreover, the pN135D-induced TCLs appeared to use a broader repertoire of T-cell receptor Vβ subtypes than those induced with p127-142. Conclusion An APL of an immunodominant allergen epitope was able to induce a novel T H 1-deviated T-cell population cross-reactive with the natural epitope in vitro . This cell population could have a therapeutic immunomodulatory function in vivo through bystander suppression. Clinical implications These results support the idea that altered peptide ligands may be used to enhance the efficacy of peptide immunotherapy.

Allergens from various sources have been shown to comprise several isoforms. In the present study, a series of chromatographic steps was carried out to separate the lipocalin allergen Bos d 2 isoforms present in cow dander. Subsequent HPLC-MS-MS analyses revealed two new Bos d 2 variants. In one of the proteins, tyrosine (Y83) was substituted by aspartic acid, and in the other protein valine (V102) was replaced by alanine. We propose the three Bos d 2 variants be named as Bos d 2.0101 (previously sequenced Bos d 2), Bos d 2.0102 and Bos d 2.0103. Our results suggest that molecular polymorphism is a common property among lipocalin allergens. Since allergen isoforms may show variation in their IgE binding and/or T-cell reactivity, all of the many allergen forms should be taken into account when planning preparations for immunotherapy.

The objective of this study was to characterize preliminarily immunogenic components characteristic of Stachybotrys chartarum to be used later for the development of a detection method for the fungus in environmental samples. The procedure for S.chartarum extract preparation was first optimized related to the age of the culture, culture type, and growth medium, and the antigenic composition of S. chartarum cultured in two different media was then characterized by the sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS–PAGE) and immunoblotting method. Cross-reactivity of S. chartarum antigenic components with 10 other fungal species was identified by the inhibition immunoblotting method. The 10-day-old S. chartarum culture extract cultured in malt extract broth revealed a wider selection of proteins and antigenic components than the 30-day-old culture extract or the culture medium extracts. When cultured in cellulose broth, S. chartarum produced a higher number of proteins and antigenic components than in malt extract broth. The most dominant immunogenic components of S. chartarum cultured in cellulose broth were those of 65, 50, 37, and 27 kDa. The components of 65 and 50 kDa proved to be the most characteristic of this fungus according to the inhibition immunoblotting analyses. Several of the S. chartarum components were identified as glycoproteins. Carbohydrate moieties of the S. chartarum components also possessed an antibody binding activity.

The term allergy refers to clinical conditions caused by an inappropriate immune response to innocuous proteins in genetically predisposed persons. Allergens of animal origin are responsible for a significant proportion of allergies. In recent years, it has become evident that practically all respiratory animal allergens characterized at the molecular level belong to the lipocalin family of proteins. The current list comprises the major allergens of horse, cow, dog, mouse, rat and cockroach as well as beta-lactoglobulin of cow's milk. While the molecular structure of all these allergens is known, far less information is available regarding their immunological characteristics. Knowing the way the immune system recognizes these allergens and reacts to them might, however, be the key for discovering the common denominator of the allergenicity of lipocalins. The human body contains numerous endogenous lipocalins, and the immune system has to adapt to their presence. We have proposed that under these conditions the immune response against the lipocalin allergens which are structurally related to endogenous lipocalins might be the pathway to allergy in genetically predisposed persons. The same might well apply also to other allergens with homologous endogenous counterparts.

The three-dimensional structure of the major bovine allergen Bos d 2 has been determined by using x-ray diffraction at 1.8-A resolution. Structurally Bos d 2 is a member of the lipocalin family comprising proteins with transport functions. There is a flat small cavity inside the Bos d 2 protein core suitable for ligand binding, and it is possible that Glu115 and Asn37 inside the core are able to make hydrogen bonds with the ligand. Many allergens from different animals belong to the lipocalin family. The amino acid residue similarities between these lipocalins indicate putative regions for IgE binding. Comparison with the available allergen structures from other sources suggests that these allergens are roughly the same size and that their shape is more spherical than elliptical.

Background: Domestic mammals are important sources of indoor allergens. However, the origin at the tissue level and the biologic function of mammalian allergens are largely unknown. Objective: The aim of this study was to localize the source of the major bovine dander allergen, Bos d 2, in bovine tissues. Methods: Samples from several organs were tested for the presence of mRNA encoding Bos d 2 and Bos d 2 protein by using the reverse transcriptase polymerase chain reaction and immunohistochemical staining, respectively. Results: Skin proved to be the only tissue where mRNA encoding Bos d 2 was detected. This observation was confirmed by immunohistochemistry with a monoclonal anti-Bos d 2 antibody as the primary antibody. In the skin sections, Bos d 2 was found in the secretory cells of apocrine sweat glands and the basement membranes of the epithelium and hair follicles. Bos d 2 belongs to the family of lipocalins comprising a number of pheromone carrier proteins that are present, for example, in the secretions of the apocrine sweat glands. Conclusion: Together with earlier data, our findings suggest that Bos d 2 is produced in sweat glands and transported to the skin surface as a carrier of the pheromone ligand. Because dander allergens of a number of mammalian species are lipocalins, the common biologic function of being pheromone carriers seems to be a common feature of an important group of aeroallergens. (J Allergy Clin Immunol 1998;101:349-53.)

More than 500 CD4 + T-cell clones (TCCs) derived from the skin of eight patients with atopic dermatitis (AD), two patients with nonatopic dermatologic disorders, two patients with allergic rhinitis, and one healthy nonatopic donor were analyzed for both their pattern of cytokine production and their antigen specificity. The proportions of TCCs from patients with AD producing interleukin-4 in response to stimulation with phorbol 12-myristate 13-acetate plus anti-CD3 antibody were higher, whereas the proportions of interferon-γ–producing TCCs were lower than those of control subjects. In two patients with AD, the majority of TCCs had a T H2 /T H0 -like phenotype, whereas in six patients with AD a T H1 /T H0 -like phenotype was prevalent. TCCs with a T H2 /T H0 -like phenotype were also isolated from the healthy skin of two patients with allergic rhinitis and one nonatopic donor. In contrast, no T H2 -like TCCs were derived from the skin of the two patients with dermatologic disorders of nonallergic origin. No unambiguous correlation was found between the proportions of TCCs producing interleukin-4 or interferon-γ (or of TCCs with T H2 - or T H1 -like profile) and the level of total serum IgE, suggesting that CD4 + T cells infiltrating the atopic skin do not play a major role in the production of serum IgE antibodies. When TCCs from five patients with AD were examined for their specificity, the proportions of allergen-specific ( Dermatophagoides pteronyssinus and Lol p 1) clones were consistently 6% or lower even in patients with high titers of ryegrass- or D. pteronyssinus –specific IgE antibodies. Because similar percentages of allergen-specific TCCs were found in skin from two healthy control subjects, the role of aeroallergens in favoring and maintaining skin lesions in patients with AD remains unclear. (J ALLERGY CLIN IMMUNOL 1995;96:411-20.)

A monoclonal antibody was developed to the 20 kd major allergen of cow by immunizing mice with crude dander extract. The monoclonal antibody did not exhibit cross-reactivity to cat, dog, and horse dander extracts when studied by ELISA inhibition. The antibody was used in affinity chromatography for the purification of the 20 kd allergen from cow dander extract. Purity of the allergen was estimated to be 88%, and allergenic reactivity was verified by IgE immunoblotting and skin prick tests. After further purification with size-exclusion chromatography, the allergen was almost 100% pure. The isoelectric point of the double-purified allergen was determined to be 4.1. The amino acid composition was characterized by the predominance of acidic amino acids.