Using active learning for monitoring networks design: The example of wind power plants sites evaluation

Locating new wind farms is of crucial importance for energy policies of the next decade. To select the new location, an accurate picture of the wind fields is necessary. However, characterizing wind fields is a difficult task, since the phenomenon is highly nonlinear and related to complex topographical features. In this paper, we propose both a nonparametric model to estimate wind speed at different time instants and a procedure to discover underrepresented topographic conditions, where new measuring stations could be added. Compared to space filling techniques, this last approach privileges optimization of the output space, thus locating new potential measuring sites through the uncertainty of the model itself.

Automatic Mallampati Classification Using Active Appearance Models

Difficult tracheal intubation assessment is an important research topic in anesthesia as failed intubations are important causes of mortality in anesthetic practice. The modified Mallampati score is widely used, alone or in conjunction with other criteria, to predict the difficulty of intubation. This work presents an automatic method to assess the modified Mallampati score from an image of a patient with the mouth wide open. For this purpose we propose an active appearance models (AAM) based method and use linear support vector machines (SVM) to select a subset of relevant features obtained using the AAM. This feature selection step proves to be essential as it improves drastically the performance of classification, which is obtained using SVM with RBF kernel and majority voting. We test our method on images of 100 patients undergoing elective surgery and achieve 97.9% accuracy in the leave-one-out crossvalidation test and provide a key element to an automatic difficult intubation assessment system.

Small ubiquitin-like modifier conjugating enzyme with active site mutation acts as dominant negative inhibitor of SUMO conjugation in arabidopsis(F).

Small ubiquitin-like modifier (SUMO) conjugation affects a broad range of processes in plants, including growth, flower initiation, pathogen defense, and responses to abiotic stress. Here, we investigate in vivo and in vitro a SUMO conjugating enzyme with a Cys to Ser change in the active site, and show that it has a dominant negative effect. In planta expression significantly perturbs normal development, leading to growth retardation, early flowering and gene expression changes. We suggest that the mutant protein can serve as a probe to investigate sumoylation, also in plants for which poor genetic infrastructure precludes analysis via loss-of-function mutants.

Kaposi sarcoma herpes virus antibody response and viremia following highly active antiretroviral therapy in the Swiss HIV Cohort study.

OBJECTIVE: To describe the effect of HAART on Kaposi sarcoma herpes virus (KSHV) antibody response and viremia among HIV-positive MSM. DESIGN: A follow-up study of 272 HIV-positive MSM (including 22 with Kaposi sarcoma) who first initiated HAART between January 1996 and July 2004 in the Swiss HIV Cohort Study. METHODS: For each individual, two serum samples, one at HAART initiation and another 24 months later, were tested for latent and lytic KSHV antibodies using immunofluorescence assays, and for KSHV viremia using PCR. Factors associated with changes in KSHV antibody titers and viremia were evaluated. RESULTS: At HAART initiation, 69.1 and 75.0% of patients were seropositive to latent and lytic KSHV antibodies, respectively. Seropositivity was associated with the presence of Kaposi sarcoma, older age, lower CD8 cell count and higher CD4/CD8 ratio. Prevalence of KSHV viremia at HAART initiation was 6.4%, being significantly higher among patients with Kaposi sarcoma (35.0%), and those with HIV viral loads 100 000 copies/ml (11.7%) or higher. At 24-month follow-up, geometric mean titers (GMTs) among KSHV seropositive patients increased and antibody seroprevalence was higher. Having Kaposi sarcoma and/or CD4 cell counts less than 50 cells/microl at HAART initiation was associated both with higher probability for antibody titers to increase (including seroconversion) and larger increases in GMTs. Only one of 17 viremic patients at HAART initiation had viremia at 24-month follow-up. CONCLUSION: HAART increases KSHV-specific humoral immune response and clearance of viremia among HIV-infected MSM, consistent with the dramatic protection offered by HAART against Kaposi sarcoma.

HCF-1 is cleaved in the active site of O-GlcNAc transferase.

Host cell factor-1 (HCF-1), a transcriptional co-regulator of human cell-cycle progression, undergoes proteolytic maturation in which any of six repeated sequences is cleaved by the nutrient-responsive glycosyltransferase, O-linked N-acetylglucosamine (O-GlcNAc) transferase (OGT). We report that the tetratricopeptide-repeat domain of O-GlcNAc transferase binds the carboxyl-terminal portion of an HCF-1 proteolytic repeat such that the cleavage region lies in the glycosyltransferase active site above uridine diphosphate-GlcNAc. The conformation is similar to that of a glycosylation-competent peptide substrate. Cleavage occurs between cysteine and glutamate residues and results in a pyroglutamate product. Conversion of the cleavage site glutamate into serine converts an HCF-1 proteolytic repeat into a glycosylation substrate. Thus, protein glycosylation and HCF-1 cleavage occur in the same active site.

Potential sources of biodynamically active natural products in Brazil

In contrast to China where vegatation is predominantly herbaceous, vegetation in Brazil is commonly arboreous. This fact may explain why Chinese drugs are usually acetate derived, while actual and potential natural therapeutic agents from Brazil are mostly shikimate derived. Only relatively few compounds isolated from Brazilian plants have been submitted to adequate pharmacological testing

Des céramiques aux hommes : étude céramique des premiers horizons fouillés sous la cathédrale Saint-Pierre de Genève (1er millénaire av. J.-C. - 40 apr. J.-C.)

Résumé Le travail présenté est basé sur l'analyse systématique des 145'157 fragments de céramique émanant de 2137 ensembles recueillis entre 1976 et 2003 lors de la fouille de la cathédrale Saint-Pierre. Datés entre le Ier millénaire av. J-C. et le XIXe siècle, ils témoignent de l'ampleur et de la richesse des activités menées au sein de ce site phare de la ville de Genève. Les particularités observées pour les six premiers horizons reconnus, qui couvrent une période comprise entre le Ier millénaire av. J.-C. et 40 apr. J.-C., liées aux structures a priori énigmatiques mises en évidence lors de la fouille, ont orienté le champ d'étude vers une compréhension fine du mobilier recueilli dans ces strates anciennes. Ainsi cadrée, l'étude s'est concentrée sur 237 complexes céramiques ayant livré 23'129 fragments. La corrélation stratigraphique menée sur les 8000 m2 de terrain fouillé a permis de porter l'analyse sur la répartition spatiale du corpus considéré. Développée également sur le plan diachronique, cette démarche a largement contribué à la définition des activités menées sur le site. Ainsi, la concentration de fragments de panses d'amphore et de vaisselle de table importée de Campanie au pied d'un tertre mis en évidence sous le choeur de la cathédrale a conduit à reconnaître l'existence d'un banquet funéraire témoignant de l'ensevelissement sous tumulus d'un aristocrate allobroge décédé entre 120 et 70 av. J.-C. Réparties sur la totalité de la surface occupée par la nef de la cathédrale et des rues la flanquant, les quelques 1024 coupes à boire de La Tène finale mises au jour ont quant à elles permis d'identifier les surfaces de gravier qu'elles scellaient comme une aire de rassemblement ayant abrité des banquets tenus périodiquement. Le maintien de ces activités rituelles et collectives jusqu'au milieu du Ier siècle de notre ère conclut l'analyse diachronique. La mise en perspective historique des données recueillies sur le terrain souligne l'importance de l'antique oppidum extremum des Allobroges, dont le sort fut en 58 av. J.-C. à l'origine de la Guerre des Gaules.

La mémoire et le temps. Mélanges offerts à Pierre Bonnard

Vingt-et-un biblistes de Suisse, de France, d'Allemagne et d'Italie se rassemblent autour d'un thème qui n'est pas seule­ment très actuel. Ce thème a été aperçu par Pierre Bonnard comme une structure fondamentale dans la théologie du Nou­veau Testament: le souvenir, l'anamnèse. Faire mémoire de Jésus pour comprendre le temps présent. La mémoire du Jésus terrestre est cette structure qui gouverne la vie et la pensée des premiers chrétiens ; elle est le lieu des combats théologiques de Paul, de Luc, de Matthieu, de Jean... Lieu de toutes les fidélités et de toutes les déviances. Vingt-et-un exégètes reprennent ce thème, pour en mesurer la fécondité dans le champ textuel qui leur est familier. Quel rôle joue au premier siècle la mémoire de l'histoire originaire, au gré du temps qui passe? Comment décrire, et apprécier théologiquement, la relecture de l'histoire de Jésus à laquelle procèdent les auteurs du Nouveau Testament? Ces contri­butions sont un hommage à Pierre Bonnard et à son oeuvre d'exégète du Nouveau Testament.

Monitoring de la problématique du cannabis en Suisse: étude sentinelle: panels 2008 = Monitoring der Cannabisproblematik in der Schweiz: Sentinella-Studie: Panels 2008

A l'heure actuelle, le monitoring de la problématique du cannabis en Suisse constitue un ensemble de travaux qui permettent le suivi de la situation au niveau national et qui sont mis en oeuvre par un consortium d'instituts. Ce monitoring comprend l'étude présentée dans ce rapport, l'étude sentinelle. Elle s'intéresse à l'évolution de la situation en matière de cannabis ainsi qu'à la gestion de cette situation au niveau local. Ainsi, les observations relevées par des professionnels de terrain dans différents domaines (santé/social, école/formation professionnelle, police/justice) et dans quatre cantons suisses (St Gall, Tessin, Vaud, Zurich), dits "sentinelle", sont récoltées et analysées annuellement.

A Survey of Active Learning Algorithms for Supervised Remote Sensing Image Classification

Defining an efficient training set is one of the most delicate phases for the success of remote sensing image classification routines. The complexity of the problem, the limited temporal and financial resources, as well as the high intraclass variance can make an algorithm fail if it is trained with a suboptimal dataset. Active learning aims at building efficient training sets by iteratively improving the model performance through sampling. A user-defined heuristic ranks the unlabeled pixels according to a function of the uncertainty of their class membership and then the user is asked to provide labels for the most uncertain pixels. This paper reviews and tests the main families of active learning algorithms: committee, large margin, and posterior probability-based. For each of them, the most recent advances in the remote sensing community are discussed and some heuristics are detailed and tested. Several challenging remote sensing scenarios are considered, including very high spatial resolution and hyperspectral image classification. Finally, guidelines for choosing the good architecture are provided for new and/or unexperienced user.