Improving gender classification accuracy in the wild

[EN]In this paper, we focus on gender recognition in challenging large scale scenarios. Firstly, we review the literature results achieved for the problem in large datasets, and select the currently hardest dataset: The Images of Groups. Secondly, we study the extraction of features from the face and its local context to improve the recognition accuracy. Diff erent descriptors, resolutions and classfii ers are studied, overcoming previous literature results, reaching an accuracy of 89.8%.

Kinship verification in the wild: the first kinship verification competition

In this paper, we present the Kinship Verification in the Wild Competition: the first kinship verification competition which is held in conjunction with the International Joint Conference on Biometrics 2014, Clearwater, Florida, USA. The key goal of this competition is to compare the performance of different methods on a new-collected dataset with the same evaluation protocol and develop the first standardized benchmark for kinship verification in the wild.

The FG 2015 kinship verification in the wild evaluation

[ES]The aim of the Kinship Verification in the Wild Evaluation (held in conjunction with the 2015 IEEE International Conference on Automatic Face and Gesture Recognition, Ljubljana, Slovenia) was to evaluate different kinship verification algorithms. For this task, two datasets were made available and three possible experimental protocols (unsupervised, image-restricted, and image-unrestricted) were designed. Five institutions submitted their results to the evaluation: (i) Politecnico di Torino, Italy; (ii) LIRIS-University of Lyon, France; (iii) Universidad de Las Palmas de Gran Canaria, Spain; (iv) Nanjing University of Aeronautics and Astronautics, China; and (v) Bar Ilan University, Israel. Most of the participants tackled the image-restricted challenge and experimental results demonstrated better kinship verification performance than the baseline methods provided by the organizers.

On using periocular biometric for gender classification in the wild

[EN]Gender information may serve to automatically modulate interaction to the user needs, among other applications. Within the Computer Vision community, gender classification (GC) has mainly been accomplished with the facial pattern. Periocular biometrics has recently attracted researchers attention with successful results in the context of identity recognition. But, there is a lack of experimental evaluation of the periocular pattern for GC in the wild. The aim of this paper is to study the performance of this specific facial area in the currently most challenging large dataset for the problem.

Production, purification, properties and application of the cellulases from a wild type strain of a yeast isolate

Kurzzusammenfassung Produktion, Reinigung, Eigenschaften und Anwendung von Cellulasen eines Wildtyp-Hefeisolates. Die effiziente Verwendung von Cellulose wird in naher Zukonft ein wichtiges Instrument zur Vermeidung einer Nahrungsmittel- und Energieknappheit werden. Deshalb haben wir uns intensiv mit Cellulasen befaßt, die aus Hefestämmen isoliert wurden. Die Fähigkeit der Cellulase-produktion eines Hefe-Stammes der Feuerwanze Pyrrhocoris apterus wurde genauer untersucht. Die systematische Stellung des Hefe-Isolates PAG1 wurde durch Sequenzierung der 18S rDNA bestimmt. Es zeigte eine nahe Verwandtschaft zu einem bereits beschriebenen Stämme der Gattung Trichosporon. Außerdem wurden die Wachstums-bedingungen für eine optimale Cellulase–Produktion bestimmt. Anschließend konnte eine der produzierten Cellulasen mit FPLC aufgereinigt und deren biochemische Eigenschaften (z.B. Substratspezifität, Temperatur optimum, optimaler pH-Wert, Einfluß von Chemikalien) untersucht werden. Eine Analyse der Abbau-Produkte zeigte, daß kristalline Cellulose und CMC zu Cellobiose, Cellulotriose, Cellulotetraose und Cellulopentaose in einem molaren Verhältnis von 32:16:8:1 umgesetezt wurden. Bei Zusatz von ?-Glykosidase aus demselben Hefestamm entstand nur Glucose und Cellobiose in einem molaren Verhältnis von 1:10. Da bisher nur eine Publikation über Cellulase-produzierende Hefe-Stämme erschienen ist, zeigen auch unsere Untersuchungen, daß Wildtyp-Hefestämme Cellulasen mit interessanten Eigenschaften produzieren können.

Evolutionary tendencies in flowers of Marantaceae with special reference to the style movement mechanism

One of the quickest plant movements ever known is made by the ´explosive´ style in Marantaceae in the service of secondary pollen presentation – herewith showing a striking apomorphy to the sister Cannaceae that might be of high evolutionary consequence. Though known already since the beginning of the 19th century the underlying mechanism of the movement has hitherto not been clarified. The present study reports about the biomechanics of the style-staminode complex and the hydraulic principles of the movement. For the first time it is shown by experiment that in Maranta noctiflora through longitudinal growth of the maturing style in the ´straitjacket´ of the hooded staminode both the hold of the style prior to its release and its tensioning for the movement are brought about. The longer the style grows in relation to the enclosing hooded staminode the more does its capacity for curling up for pollen transfer increase. Hereby I distinguish between the ´basic tension´ that a growing style builds up anyway, even when the hooded staminode is removed beforehand, and the ´induced tension´ which comes about only under the pressure of a ´too short´ hooded staminode and which enables the movement. The results of these investigations are discussed in view of previous interpretations ranging from possible biomechanical to electrophysiological mechanisms. To understand furthermore by which means the style gives way to the strong bending movement without suffering outwardly visible damage I examined its anatomical structure in several genera for its mechanical and hydraulic properties and for the determination of the entire curvature after release. The actual bending part contains tubulate cells whose walls are extraordinarily porous and large longitudinal intercellular spaces. SEM indicates the starting points of cell-wall loosening in primary walls and lysis of middle lamellae - probably through an intense pectinase activity in the maturing style. Fluorescence pictures of macerated and living style-tissue confirm cell-wall perforations that do apparently connect neighbouring cells, which leads to an extremely permeable parenchyma. The ´water-body´ can be shifted from central to dorsal cell layers to support the bending. The geometrical form of the curvature is determined by the vascular bundles. I conclude that the style in Marantaceae contains no ´antagonistic´ motile tissues as in Mimosa or Dionaea. Instead, through self-maceration it develops to a ´hydraulic tissue´ which carries out an irreversible movement through a sudden reshaping. To ascertain the evolutionary consequence of this apomorphic pollination mechanism the diversity and systematic value of hooded staminodes are examined. For this hooded staminodes of 24 genera are sorted according to a minimalistic selection of shape characters and eight morphological types are abstracted from the resulting groups. These types are mapped onto an already available maximally parsimonious tree comprising five major clades. An amazing correspondence is found between the morphological types and the clades; several sister-relationships are confirmed and in cases of uncertain position possible evolutionary pathways, such as convergence, dispersal or re-migration, are discussed, as well as the great evolutionary tendencies for the entire family in which – at least as regards the shape of hooded staminodes – there is obviously a tendency from complicated to strongly simplified forms. It suggests itself that such simplifying derivations may very likely have taken place as adaptations to pollinating animals about which at present too little is known. The value of morphological characters in relation to modern phylogenetic analysis is discussed and conditions for the selection of morphological characters valuable for a systematic grouping are proposed. Altogether, in view of the evolutionary success of Marantaceae compared with Cannaceae the movement mechanism of the style-staminode complex can safely be considered a key innovation within the order Zingiberales.

Variable Na(v)1.5 protein expression from the wild-type allele correlates with the penetrance of cardiac conduction disease in the Scn5a(+/-) mouse model

BACKGROUND: Loss-of-function mutations in SCN5A, the gene encoding Na(v)1.5 Na+ channel, are associated with inherited cardiac conduction defects and Brugada syndrome, which both exhibit variable phenotypic penetrance of conduction defects. We investigated the mechanisms of this heterogeneity in a mouse model with heterozygous targeted disruption of Scn5a (Scn5a(+/-) mice) and compared our results to those obtained in patients with loss-of-function mutations in SCN5A. METHODOLOGY/PRINCIPAL FINDINGS: Based on ECG, 10-week-old Scn5a(+/-) mice were divided into 2 subgroups, one displaying severe ventricular conduction defects (QRS interval>18 ms) and one a mild phenotype (QRS< or = 18 ms; QRS in wild-type littermates: 10-18 ms). Phenotypic difference persisted with aging. At 10 weeks, the Na+ channel blocker ajmaline prolonged QRS interval similarly in both groups of Scn5a(+/-) mice. In contrast, in old mice (>53 weeks), ajmaline effect was larger in the severely affected subgroup. These data matched the clinical observations on patients with SCN5A loss-of-function mutations with either severe or mild conduction defects. Ventricular tachycardia developed in 5/10 old severely affected Scn5a(+/-) mice but not in mildly affected ones. Correspondingly, symptomatic SCN5A-mutated Brugada patients had more severe conduction defects than asymptomatic patients. Old severely affected Scn5a(+/-) mice but not mildly affected ones showed extensive cardiac fibrosis. Mildly affected Scn5a(+/-) mice had similar Na(v)1.5 mRNA but higher Na(v)1.5 protein expression, and moderately larger I(Na) current than severely affected Scn5a(+/-) mice. As a consequence, action potential upstroke velocity was more decreased in severely affected Scn5a(+/-) mice than in mildly affected ones. CONCLUSIONS: Scn5a(+/-) mice show similar phenotypic heterogeneity as SCN5A-mutated patients. In Scn5a(+/-) mice, phenotype severity correlates with wild-type Na(v)1.5 protein expression.

Regulation of wingless-type MMTV integration site family (WNT) signalling in pancreatic islets from wild-type and obese mice

TCF7L2 is a type 2 diabetes susceptibility gene and downstream effector of canonical wingless-type MMTV integration site family (WNT) signalling. However, it is unknown whether this pathway is active in adult pancreatic islets in vivo, and whether it is regulated in obesity.

No evidence for large differences in genomic methylation between wild and hatchery steelhead (Oncorhynchus mykiss)

When salmonid fish that have been raised in hatcheries spawn in the wild, they often produce fewer surviving adult offspring than wild fish. Recent data from steelhead (Oncorhynchus mykiss) in the Hood River (Oregon, USA) show that even one or two generations of hatchery culture can result in dramatic declines in fitness. Although intense domestication selection could cause such declines, it is worth considering alternative explanations. One possibility is heritable epigenetic changes induced by the hatchery environment. Here, we show, using methylation-sensitive amplified fragment length polymorphism, that hatchery and wild adult steelhead from the Hood River do not appear to differ substantially in overall levels of genomic methylation. Thus, although altered methylation of specific DNA sites or other epigenetic processes could still be important, the hatchery environment does not appear to cause a global hypo- or hypermethylation of the genome or create a large number of sites that are differentially methylated.

A novel steroidal antiandrogen targeting wild type and mutant androgen receptors

Prostate cancer (PCa) progression is enhanced by androgen and treatment with antiandrogens represents an alternative to castration. While patients initially respond favorably to androgen ablation therapy, most experience a relapse of the disease within 1-2 years by expressing androgen receptor (AR) mutants. Such mutations, indeed, promote unfavorable agonistic behavior from classical antagonists. Here, we have synthesized and screened 37 novel compounds derived from dihydrotestosterone (DHT), cyanolutamide and hydroxyflutamide. These derivatives were tested for their potential antagonistic activity using a luciferase reporter gene assay and binding properties were determined for wild type (WT) and mutant ARs (T877A, W741C, W741L, H874Y). In the absence and presence of antiandrogens, androgen dependent cellular proliferation and prostate specific antigen (PSA) expression were assayed in the prostate cancer cell line LNCaP by crystal violet, real time PCR and by Western blots. Also, cellular proliferation and PSA expression were assayed in 22Rv1. A novel compound RB346, derived from DHT, was found to be an antagonist for all tested AR forms, preventing DHT induced proliferation and PSA expression in LNCaP and 22Rv1 cells. RB346 displayed no agonistic activity, in contrast to the non-steroidal antiandrogen bicalutamide (Casodex) with unfavorable agonistic activity for W741L-AR. Additionally, RB346 has a slightly higher binding affinity for WT-AR, T877A-AR and H874Y-AR than bicalutamide. Thus, RB346 is the first potent steroidal antiandrogen with efficacy for WT and various AR mutants.

A summary of the 2004 Ohio Music Education Association research forum presented by Dr. Patricia J. Flowers.

Each year, the Research Committee of the Ohio Music Education Association sponsors a half-day Research Forum prior to the beginning of the state music education association conference. In 2004, Dr. Patricia J. Flowers, Professor of Music at the Ohio State University was the guest speaker. This article summarizes her talk on the process of becoming a music education researcher