HIV rebounds from latently infected cells, rather than from continuing low-level replication.

Rapid rebound of plasma viremia in patients after interruption of long-term combination antiretroviral therapy (cART) suggests persistence of low-level replicating cells or rapid reactivation of latently infected cells. To further characterize rebounding virus, we performed extensive longitudinal clonal evolutionary studies of HIV env C2-V3-C3 regions and exploited the temporal relationships of rebounding plasma viruses with regard to pretreatment sequences in 20 chronically HIV-1-infected patients having undergone multiple 2-week structured treatment interruptions (STI). Rebounding virus during the short STI was homogeneous, suggesting mono- or oligoclonal origin during reactivation. No evidence for a temporal structure of rebounding virus in regard to pretreatment sequences was found. Furthermore, expansion of distinct lineages at different STI cycles emerged. Together, these findings imply stochastic reactivation of different clones from long-lived latently infected cells rather than expansion of viral populations replicating at low levels. After treatment was stopped, diversity increased steadily, but pretreatment diversity was, on average, achieved only >2.5 years after the start of STI when marked divergence from preexisting quasispecies also emerged. In summary, our results argue against persistence of ongoing low-level replication in patients on suppressive cART. Furthermore, a prolonged delay in restoration of pretreatment viral diversity after treatment interruption demonstrates a surprisingly sustained evolutionary bottleneck induced by punctuated antiretroviral therapy.

On Recurrent and Deep Neural Networks

L'apprentissage profond est un domaine de recherche en forte croissance en apprentissage automatique qui est parvenu à des résultats impressionnants dans différentes tâches allant de la classification d'images à la parole, en passant par la modélisation du langage. Les réseaux de neurones récurrents, une sous-classe d'architecture profonde, s'avèrent particulièrement prometteurs. Les réseaux récurrents peuvent capter la structure temporelle dans les données. Ils ont potentiellement la capacité d'apprendre des corrélations entre des événements éloignés dans le temps et d'emmagasiner indéfiniment des informations dans leur mémoire interne. Dans ce travail, nous tentons d'abord de comprendre pourquoi la profondeur est utile. Similairement à d'autres travaux de la littérature, nos résultats démontrent que les modèles profonds peuvent être plus efficaces pour représenter certaines familles de fonctions comparativement aux modèles peu profonds. Contrairement à ces travaux, nous effectuons notre analyse théorique sur des réseaux profonds acycliques munis de fonctions d'activation linéaires par parties, puisque ce type de modèle est actuellement l'état de l'art dans différentes tâches de classification. La deuxième partie de cette thèse porte sur le processus d'apprentissage. Nous analysons quelques techniques d'optimisation proposées récemment, telles l'optimisation Hessian free, la descente de gradient naturel et la descente des sous-espaces de Krylov. Nous proposons le cadre théorique des méthodes à région de confiance généralisées et nous montrons que plusieurs de ces algorithmes développés récemment peuvent être vus dans cette perspective. Nous argumentons que certains membres de cette famille d'approches peuvent être mieux adaptés que d'autres à l'optimisation non convexe. La dernière partie de ce document se concentre sur les réseaux de neurones récurrents. Nous étudions d'abord le concept de mémoire et tentons de répondre aux questions suivantes: Les réseaux récurrents peuvent-ils démontrer une mémoire sans limite? Ce comportement peut-il être appris? Nous montrons que cela est possible si des indices sont fournis durant l'apprentissage. Ensuite, nous explorons deux problèmes spécifiques à l'entraînement des réseaux récurrents, à savoir la dissipation et l'explosion du gradient. Notre analyse se termine par une solution au problème d'explosion du gradient qui implique de borner la norme du gradient. Nous proposons également un terme de régularisation conçu spécifiquement pour réduire le problème de dissipation du gradient. Sur un ensemble de données synthétique, nous montrons empiriquement que ces mécanismes peuvent permettre aux réseaux récurrents d'apprendre de façon autonome à mémoriser des informations pour une période de temps indéfinie. Finalement, nous explorons la notion de profondeur dans les réseaux de neurones récurrents. Comparativement aux réseaux acycliques, la définition de profondeur dans les réseaux récurrents est souvent ambiguë. Nous proposons différentes façons d'ajouter de la profondeur dans les réseaux récurrents et nous évaluons empiriquement ces propositions.

Assessing the sources and magnitude of diurnal nitrate variability in the San Joaquin River (California) with an in-situ optical nitrate sensor and dual nitrate isotopes

We investigated diurnal nitrate (NO3-) concentration variability in the San Joaquin River using an in situ optical NO3- sensor and discrete sampling during a 5-day summer period characterized by high algal productivity. Dual NO3- isotopes (delta N-15(NO3) and delta O-18(NO3)) and dissolved oxygen isotopes (delta O-18(DO)) were measured over 2 days to assess NO3- sources and biogeochemical controls over diurnal time-scales. Concerted temporal patterns of dissolved oxygen (DO) concentrations and delta O-18(DO) were consistent with photosynthesis, respiration and atmospheric O-2 exchange, providing evidence of diurnal biological processes independent of river discharge. Surface water NO3- concentrations varied by up to 22% over a single diurnal cycle and up to 31% over the 5-day study, but did not reveal concerted diurnal patterns at a frequency comparable to DO concentrations. The decoupling of delta N-15(NO3) and delta O-18(NO3) isotopes suggests that algal assimilation and denitrification are not major processes controlling diurnal NO3- variability in the San Joaquin River during the study. The lack of a clear explanation for NO3- variability likely reflects a combination of riverine biological processes and time-varying physical transport of NO3- from upstream agricultural drains to the mainstem San Joaquin River. The application of an in situ optical NO3- sensor along with discrete samples provides a view into the fine temporal structure of hydrochemical data and may allow for greater accuracy in pollution assessment.

Simulations of the London urban heat island

We present simulations of London's meteorology using the Met Office Unified Model with a new, sophisticated surface energy-balance scheme to represent the urban surfaces, called MORUSES. Simulations are performed with the urban surfaces represented and with the urban surfaces replaced with grass in order to calculate the urban increment on the local meteorology. The local urban effects were moderated to some extent by the passage of an onshore flow that propagated up the Thames estuary and across the city, cooling London slightly in the afternoon. Validations of screen-level temperature show encouraging agreement to within 1–2 K, when the urban increment is up to 5 K. The model results are then used to examine factors shaping the spatial and temporal structure of London's atmospheric boundary layer. The simulations reconcile the differences in the temporal evolution of the urban heat island (UHI) shown in various studies and demonstrate that the variation of UHI with time depends strongly on the urban fetch. The UHI at a location downwind of the city centre shows a decrease in UHI during the night, while the UHI at the city centre stays constant. Finally, the UHI at a location upwind of the city centre increases continuously. The magnitude of the UHI by the time of the evening transition increases with urban fetch. The urban increments are largest at night, when the boundary layer is shallow. The boundary layer experiences continued warming after sunset, as the heat from the urban fabric is released, and a weakly convective boundary layer develops across the city. The urban land-use fraction is the dominant control on the spatial structure in the sensible heat flux and the resulting urban increment, although even the weak advection present in this case study is sufficient to advect the peak temperature increments downwind of the most built-up areas. Copyright © 2011 Royal Meteorological Society and British Crown Copyright, the Met Office

Realism of rainfall in a very high-resolution regional climate model

The realistic representation of rainfall on the local scale in climate models remains a key challenge. Realism encompasses the full spatial and temporal structure of rainfall, and is a key indicator of model skill in representing the underlying processes. In particular, if rainfall is more realistic in a climate model, there is greater confidence in its projections of future change. In this study, the realism of rainfall in a very high-resolution (1.5 km) regional climate model (RCM) is compared to a coarser-resolution 12-km RCM. This is the first time a convection-permitting model has been run for an extended period (1989–2008) over a region of the United Kingdom, allowing the characteristics of rainfall to be evaluated in a climatological sense. In particular, the duration and spatial extent of hourly rainfall across the southern United Kingdom is examined, with a key focus on heavy rainfall. Rainfall in the 1.5-km RCM is found to be much more realistic than in the 12-km RCM. In the 12-km RCM, heavy rain events are not heavy enough, and tend to be too persistent and widespread. While the 1.5-km model does have a tendency for heavy rain to be too intense, it still gives a much better representation of its duration and spatial extent. Long-standing problems in climate models, such as the tendency for too much persistent light rain and errors in the diurnal cycle, are also considerably reduced in the 1.5-km RCM. Biases in the 12-km RCM appear to be linked to deficiencies in the representation of convection.

Stratosphere‐troposphere coupling and annular mode variability in chemistry‐climate models

The internal variability and coupling between the stratosphere and troposphere in CCMVal‐2 chemistry‐climate models are evaluated through analysis of the annular mode patterns of variability. Computation of the annular modes in long data sets with secular trends requires refinement of the standard definition of the annular mode, and a more robust procedure that allows for slowly varying trends is established and verified. The spatial and temporal structure of the models’ annular modes is then compared with that of reanalyses. As a whole, the models capture the key features of observed intraseasonal variability, including the sharp vertical gradients in structure between stratosphere and troposphere, the asymmetries in the seasonal cycle between the Northern and Southern hemispheres, and the coupling between the polar stratospheric vortices and tropospheric midlatitude jets. It is also found that the annular mode variability changes little in time throughout simulations of the 21st century. There are, however, both common biases and significant differences in performance in the models. In the troposphere, the annular mode in models is generally too persistent, particularly in the Southern Hemisphere summer, a bias similar to that found in CMIP3 coupled climate models. In the stratosphere, the periods of peak variance and coupling with the troposphere are delayed by about a month in both hemispheres. The relationship between increased variability of the stratosphere and increased persistence in the troposphere suggests that some tropospheric biases may be related to stratospheric biases and that a well‐simulated stratosphere can improve simulation of tropospheric intraseasonal variability.

A dynamic causal model of the coupling between pulse stimulation and neural activity

We present a dynamic causal model that can explain context-dependent changes in neural responses, in the rat barrel cortex, to an electrical whisker stimulation at different frequencies. Neural responses were measured in terms of local field potentials. These were converted into current source density (CSD) data, and the time series of the CSD sink was extracted to provide a time series response train. The model structure consists of three layers (approximating the responses from the brain stem to the thalamus and then the barrel cortex), and the latter two layers contain nonlinearly coupled modules of linear second-order dynamic systems. The interaction of these modules forms a nonlinear regulatory system that determines the temporal structure of the neural response amplitude for the thalamic and cortical layers. The model is based on the measured population dynamics of neurons rather than the dynamics of a single neuron and was evaluated against CSD data from experiments with varying stimulation frequency (1–40 Hz), random pulse trains, and awake and anesthetized animals. The model parameters obtained by optimization for different physiological conditions (anesthetized or awake) were significantly different. Following Friston, Mechelli, Turner, and Price (2000), this work is part of a formal mathematical system currently being developed (Zheng et al., 2005) that links stimulation to the blood oxygen level dependent (BOLD) functional magnetic resonance imaging (fMRI) signal through neural activity and hemodynamic variables. The importance of the model described here is that it can be used to invert the hemodynamic measurements of changes in blood flow to estimate the underlying neural activity.

ReadingAct RGB-D action dataset and human action recognition from local features

For general home monitoring, a system should automatically interpret people’s actions. The system should be non-intrusive, and able to deal with a cluttered background, and loose clothes. An approach based on spatio-temporal local features and a Bag-of-Words (BoW) model is proposed for single-person action recognition from combined intensity and depth images. To restore the temporal structure lost in the traditional BoW method, a dynamic time alignment technique with temporal binning is applied in this work, which has not been previously implemented in the literature for human action recognition on depth imagery. A novel human action dataset with depth data has been created using two Microsoft Kinect sensors. The ReadingAct dataset contains 20 subjects and 19 actions for a total of 2340 videos. To investigate the effect of using depth images and the proposed method, testing was conducted on three depth datasets, and the proposed method was compared to traditional Bag-of-Words methods. Results showed that the proposed method improves recognition accuracy when adding depth to the conventional intensity data, and has advantages when dealing with long actions.

The SPARC Data Initiative: comparisons of CFC-11, CFC-12, HF and SF6 climatologies from international satellite limb sounders

A quality assessment of the CFC-11 (CCl3F), CFC-12 (CCl2F2), HF, and SF6 products from limb-viewing satellite instruments is provided by means of a detailed intercomparison. The climatologies in the form of monthly zonal mean time series are obtained from HALOE, MIPAS, ACE-FTS, and HIRDLS within the time period 1991–2010. The intercomparisons focus on the mean biases of the monthly and annual zonal mean fields and aim to identify their vertical, latitudinal and temporal structure. The CFC evaluations (based on MIPAS, ACE-FTS and HIRDLS) reveal that the uncertainty in our knowledge of the atmospheric CFC-11 and CFC-12 mean state, as given by satellite data sets, is smallest in the tropics and mid-latitudes at altitudes below 50 and 20 hPa, respectively, with a 1σ multi-instrument spread of up to ±5 %. For HF, the situation is reversed. The two available data sets (HALOE and ACE-FTS) agree well above 100 hPa, with a spread in this region of ±5 to ±10 %, while at altitudes below 100 hPa the HF annual mean state is less well known, with a spread ±30 % and larger. The atmospheric SF6 annual mean states derived from two satellite data sets (MIPAS and ACE-FTS) show only very small differences with a spread of less than ±5 % and often below ±2.5 %. While the overall agreement among the climatological data sets is very good for large parts of the upper troposphere and lower stratosphere (CFCs, SF6) or middle stratosphere (HF), individual discrepancies have been identified. Pronounced deviations between the instrument climatologies exist for particular atmospheric regions which differ from gas to gas. Notable features are differently shaped isopleths in the subtropics, deviations in the vertical gradients in the lower stratosphere and in the meridional gradients in the upper troposphere, and inconsistencies in the seasonal cycle. Additionally, long-term drifts between the instruments have been identified for the CFC-11 and CFC-12 time series. The evaluations as a whole provide guidance on what data sets are the most reliable for applications such as studies of atmospheric transport and variability, model–measurement comparisons and detection of long-term trends. The data sets will be publicly available from the SPARC Data Centre and through PANGAEA (doi:10.1594/PANGAEA.849223).

Bioacoustic analysis of advertisement call in Hyla nana and Hyla sanborni (Anura, Hylidae) in Botucatu, São Paulo, Brazil

As características relacionadas à comunicação sonora em duas espécies aparentadas, Hyla nana e H. sanborni, pertencentes ao grupo nana de espécies, foram estudadas, entre agosto de 1997 e junho de 1999, em duas poças permanentes de grande porte em área aberta na região de Botucatu, Estado de São Paulo. Foram obtidas gravações de 131 exemplares, 71 indivíduos de H. nana e 58 de H. sanborni, durante início de atividade de vocalização e atividade de vocalização em coro. Houve diferença nos ritmos de emissão de notas dos cantos entre o início das atividades e durante os coros. O canto de anúncio das espécies consiste na emissão de séries consecutivas de notas simples, pulsadas, com taxa de repetição rápida. Hyla nana e H. sanborni apresentam dois tipos de notas em seu canto de anúncio, denominados aqui como tipos A e B. Notas do tipo A, introdutórias, apresentam maior duração e número de pulsos, e suas emissões foram mais freqüentes durante o início das atividades de vocalização. As notas introdutórias são as primeiras da série emitida em atividade de coro. As notas do tipo B, secundárias, são curtas e com menor número de pulsos, sendo emitidas durante as vocalizações em coro. Os dois tipos de notas encontrados diferem significativamente em sua estrutura temporal. As duas espécies apresentaram segregação acústica tanto na estrutura espectral como na temporal de seus cantos de anúncio.

Canto de anúncio de Hyla jimi e Hyla elianeae (Anura, Hylidae) na região de Botucatu, São Paulo, Brasil

In this work we present a description of the advertisement call of H. jimi, the locality type, and of H. elianeae, until now unknown in the literature. Field activities were carried out from August 1997 through June 1999; recordings were made in two open-area environments in the Botucatu region, São Paulo State. Vocalizations of 100 individuals (49 H. jimi and 51 H. elianeae) were recorded; nine characteristics of the advertisement call were examined. The advertisement calls of both species consist of consecutive series of simple notes with relatively fast repetition rates. Males of H. jimi and H. elianeae presented two patterns of note emission: one emitted by individuals beginning vocalization activity or isolated from the aggregate, and another emitted by males in chorus activity interacting with closely neighboring males. A significant difference was verified in the temporal structure of the two vocalization patterns.

Calling sites and acoustic partitioning in species of the Hyla nana and rubicundula groups (Anura, Hylidae)

We analysed spatial and acoustic partitioning among four species of Hyla belonging to two species-groups: nana (H. nana and H. sanborni) and rubicundula (H. elianeae and H. jimi). Field activities were conducted at three permanent ponds, from 1998 through 2001. Four attributes of the calling sites were analysed: perch height, distance of the perch from the edge of the pond, type of perch (vegetation) and the individual's position on the perch. There was extensive overlap in the four calling-site variables analysed. However, we found spatial segregation did occur in calling site height and the distance of perches from pond edges. Bioacoustic analyses revealed behavioural differences among species in calling activity, both time of onset and peak calling in chorus. There was acoustic partitioning among species the fundamental frequency of the advertisement calls, principally as a function of the temporal structure (e.g. note duration, rate of note repetition, duration and rate of repetition of the calling pulses). We propose that differences in physical attributes of calling site and in characteristics of calls allow these species to exist in sympatry.

Psicologia, metafísica e literatura: A descrição dos sentimentos profundos em Bergson

This paper aims to explore the relation between psychology, metaphysics and literature, through an examination of Bergon's Time and Free Will: An Essay on the Immediate Data of Consciousness, or, more precisely, through the description of deep feelings, which represent in the Essay a privileged moment for understanding the temporal structure of consciousness. However, this study will not be restricted only to Bergson's text and its descriptions of deep feelings (such as aesthetic and moral emotions), which would probably be repetitive. Instead, we shall use a work of literature (Guimarães Rosa's novel, Grande sertão: veredas) to exemplify the possibility of a qualitative description of stream of consciousness, revealing its temporal structure. We hope in this way to clarify the interaction between psychology, metaphysics and literature in the philosophy of Bergson.

Composition of functional ecological guilds of the fish fauna of the internal sector of the Amazon Estuary, Pará, Brazil

O presente estudo descreve a estrutura espacial e temporal da comunidade de peixes estuarinos da porção interna do estuário Amazônico. Amostras foram obtidas no canal principal e canais de maré das Baías do Guajará e Marajó e rio Guamá. Foram coletados um total de 41.516 espécimes, correspondendo 136 espécies, 38 famílias e 12 ordens. Na estação seca, a salinidade média no canal principal aumentou ao longo do gradiente limnico–marinho, entre o rio Guamá e a Baía do Marajó. A riqueza de espécies foi mais baixa na foz do rio Guamá e na margem direita da baía do Guajará. A composição das espécies e as guildas ambientais diferiram significativamente entre as áreas: Migrantes e Ocasionais de água doce foram dominantes no rio Guamá e na Baía do Guajará, enquanto Estuarinas, Marinhas Migrantes e Ocasionais dominaram na Baía do Marajó. Contudo, as guildas tróficas foram relativamente bem balanceadas, em termos funcionais. Piscívoros e Zoobentívoros foram os grupos alimentares dominantes em todas as áreas. Neste estudo, a avaliação da comunidade e o uso da abordagem com guildas foram eficientes para descrever a estrutura e o funcionamento das assembleias de peixes estuarinos e também como ferramenta na avaliação das pressões antrópicas na área.

Distribuição, dieta e ecomorfologia das espécies de peixes no sistema do Ribeirão Grande, no município de Pindamonhangaba, SP

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)