Patterns of frugivory in the columnar cactus Pilosocereus leucocephalus

In the frugivory networks of many arid and semi‐arid Mesoamerican ecosystems, columnar cacti act as keystone species that produce fruits with a high content of water and nutrients attractive to numerous vertebrates. The aim of this investigation was to assess the fruit removal patterns of two guilds of frugivores on the fruits of the woolly torch Pilosocereus leucocephalus. We assessed fruit pulp removal in two ways: by estimating the consumption of seeds given the amount of pulp removed per visit and by estimating the percentage of pulp removal over time. We put exclosures on unripe, intact fruits to keep frugivores from removing material. Once ripe, we removed the exclosures and tracked animal visitation of 69 fruits using camera traps. We obtained a total of 2,162 hr of footage (14:47 hours of them with effective pulp removal). The highest number of visitors is that of diurnal species (n = 12, all birds) versus only four nocturnal (three bats, one rodent). The most effective species in pulp removal are birds. Bats play a modest role in frugivory of this cactus. The significance of this work is twofold: (a) birds and bats consume the fruit pulp of this cactus and likely disperse its seeds, and (b) although bats rank high in pulp removal effectiveness, birds as a guild far outweigh their importance in this system, as they are not only more frequent but also remove more pulp and seeds. Both groups are known to be important in cacti seed dispersal, and our findings are essential in understanding the population dynamics of the woolly torch and in elucidating its seed dispersal ecology.     

PLS-ANN判别分析自体荧光光谱识别胃癌

本文对58例胃癌病人离体标本的癌浆膜和正常浆膜进行以308nm为激发光的自体荧光光谱检测,采用多因素分析法进行光谱信息提取,以识别胃癌。研究表明偏最小二乘法结合神经网络法（简称PLS—ANN）进行判别分析,诊断胃癌的灵敏度为86％,特异度为100％,准确率为93％,有望成为手术中快速识别胃癌在胃壁的浸润范围的有效方法。     

The spinal cord in vitro: What can it tell us about nociception?

The use of amphibian and mammalian in vitro spinal cord preparations, e.g., hemisected cord and transverse slices, has gained in popularity over the years due to the flexibility and ease of use of such preparations compared to classical in vivo approaches. When combined with modern experimental methodologies, such as patch clamping of visualized single cells or post-experimental neuroanatomy, this approach provides a powerful addition to the armamentarium available to study nociceptive processing in the dorsal horn. Some of these novel experimental approaches and the insight into nociception that they have provided are described below. Neirofiziologiya/Neurophysiology, Vol. 38, Nos. 5/6, pp. 481–491, September–December, 2006.     

The role of cross-border transhumance in influencing resident herders’ cattle husbandry practices and use of genetic resources

For centuries, the humid West African coastal country of Benin attracts cattle herders from neighboring Sahelian countries such as Niger, Burkina Faso and Nigeria. Each year, several thousands of cattle are trekked over hundreds of kilometers to Benin following established and non-established transhumance corridors. This cross-border mobility has been shown to play an important role in ensuring the productivity of the transhumant herds through adequate late dry season access to pastoral resources. Yet, its effects on the traditional agro-pastoral production systems that are encountered along the routes in Benin have never been explicitly investigated. Therefore, we collected socio-economic household data, herd characteristics and management data from 104 resident herders and 38 transhumant herders in 2 vegetation zones of Benin. To determine whether or not the proximity to transhumance corridors affects sedentary production systems, characteristics of cattle herds and farmers’ management practices were compared between villages close to (within a 25 km buffer) and far from (outside a 25 km buffer) transhumance corridors within and between vegetation zones using non-parametric statistical tests. Existing relationships between resident and transhumant herders were also identified and characterized. Subsequently, herd characteristics and management practices were compared between resident herders having relationships with transhumant herders and those without. Herd sizes of resident herders living close to transhumance corridors were larger (P < 0.01) than those of their counterparts living far away. Also, proximity to transhumance corridors had positive effects on herd management practices. The relationships between resident and transhumant herders were governed by a variety of interests including encampment/manuring contracts, exploitation of grazing lands and watering points, trading and bartering of cattle. This exchange of cattle is an important driver of change in the breed composition of local herds and represents an opportunity for resident herders to enhance their herds’ productivity through crossbreeding. However, the mere replacement or indiscriminate crossbreeding of local cattle breeds with those kept by transhumant herders threatens the sustainability of the traditional resident herding systems by increasing the risk of genetic erosion and loss of valuable adaptive traits in indigenous animal genetic resources.     

Mixed company: a framework for understanding the composition and organization of mixed‐species animal groups

Mixed‐species animal groups (MSGs) are widely acknowledged to increase predator avoidance and foraging efficiency, among other benefits, and thereby increase participants' fitness. Diversity in MSG composition ranges from two to 70 species of very similar or completely different phenotypes. Yet consistency in organization is also observable in that one or a few species usually have disproportionate importance for MSG formation and/or maintenance. We propose a two‐dimensional framework for understanding this diversity and consistency, concentrating on the types of interactions possible between two individuals, usually of different species. One axis represents the similarity of benefit types traded between the individuals, while the second axis expresses asymmetry in the relative amount of benefits/costs accrued. Considering benefit types, one extreme represents the case of single‐species groups wherein all individuals obtain the same supplementary, group‐size‐related benefits, and the other extreme comprises associations of very different, but complementary species (e.g. one partner creates access to food while the other provides vigilance). The relevance of social information and the matching of activities (e.g. speed of movement) are highest for relationships on the supplementary side of this axis, but so is competition; relationships between species will occur at points along this gradient where the benefits outweigh the costs. Considering benefit amounts given or received, extreme asymmetry occurs when one species is exclusively a benefit provider and the other a benefit user. Within this parameter space, some MSG systems are constrained to one kind of interaction, such as shoals of fish of similar species or leader–follower interactions in fish and other taxa. Other MSGs, such as terrestrial bird flocks, can simultaneously include a variety of supplementary and complementary interactions. We review the benefits that species obtain across the diversity of MSG types, and argue that the degree and nature of asymmetry between benefit providers and users should be measured and not just assumed. We then discuss evolutionary shifts in MSG types, focusing on drivers towards similarity in group composition, and selection on benefit providers to enhance the benefits they can receive from other species. Finally, we conclude by considering how individual and collective behaviour in MSGs may influence both the structure and processes of communities.     

1D-CADCapsNet: One dimensional deep capsule networks for coronary artery disease detection using ECG signals

PurposeCardiovascular disease (CVD) is a leading cause of death globally. Electrocardiogram (ECG), which records the electrical activity of the heart, has been used for the diagnosis of CVD. The automated and robust detection of CVD from ECG signals plays a significant role for early and accurate clinical diagnosis. The purpose of this study is to provide automated detection of coronary artery disease (CAD) from ECG signals using capsule networks (CapsNet).MethodsDeep learning-based approaches have become increasingly popular in computer aided diagnosis systems. Capsule networks are one of the new promising approaches in the field of deep learning. In this study, we used 1D version of CapsNet for the automated detection of coronary artery disease (CAD) on two second (95,300) and five second-long (38,120) ECG segments. These segments are obtained from 40 normal and 7 CAD subjects. In the experimental studies, 5-fold cross validation technique is employed to evaluate performance of the model.ResultsThe proposed model, which is named as 1D-CADCapsNet, yielded a promising 5-fold diagnosis accuracy of 99.44% and 98.62% for two- and five-second ECG signal groups, respectively. We have obtained the highest performance results using 2 s ECG segment than the state-of-art studies reported in the literature.Conclusions1D-CADCapsNet model automatically learns the pertinent representations from raw ECG data without using any hand-crafted technique and can be used as a fast and accurate diagnostic tool to help cardiologists.     

Predictor species: Improving assessments of rare species occurrence by modeling environmental co‐responses

Designing an effective conservation strategy requires understanding where rare species are located. Because rare species can be difficult to find, ecologists often identify other species called conservation surrogates that can help inform the distribution of rare species. Species distribution models typically rely on environmental data when predicting the occurrence of species, neglecting the effect of species' co‐occurrences and biotic interactions. Here, we present a new approach that uses Bayesian networks to improve predictions by modeling environmental co‐responses among species. For species from a European peat bog community, our approach consistently performs better than single‐species models and better than conventional multi‐species approaches that include the presence of nontarget species as additional independent variables in regression models. Our approach performs particularly well with rare species and when calibration data are limited. Furthermore, we identify a group of “predictor species” that are relatively common, insensitive to the presence of other species, and can be used to improve occurrence predictions of rare species. Predictor species are distinct from other categories of conservation surrogates such as umbrella or indicator species, which motivates focused data collection of predictor species to enhance conservation practices.     

Three critical factors affecting automated image species recognition performance for camera traps

Ecological camera traps are increasingly used by wildlife biologists to unobtrusively monitor an ecosystems animal population. However, manual inspection of the images produced is expensive, laborious, and time‐consuming. The success of deep learning systems using camera trap images has been previously explored in preliminary stages. These studies, however, are lacking in their practicality. They are primarily focused on extremely large datasets, often millions of images, and there is little to no focus on performance when tasked with species identification in new locations not seen during training. Our goal was to test the capabilities of deep learning systems trained on camera trap images using modestly sized training data, compare performance when considering unseen background locations, and quantify the gradient of lower bound performance to provide a guideline of data requirements in correspondence to performance expectations. We use a dataset provided by Parks Canada containing 47,279 images collected from 36 unique geographic locations across multiple environments. Images represent 55 animal species and human activity with high‐class imbalance. We trained, tested, and compared the capabilities of six deep learning computer vision networks using transfer learning and image augmentation: DenseNet201, Inception‐ResNet‐V3, InceptionV3, NASNetMobile, MobileNetV2, and Xception. We compare overall performance on “trained” locations where DenseNet201 performed best with 95.6% top‐1 accuracy showing promise for deep learning methods for smaller scale research efforts. Using trained locations, classifications with <500 images had low and highly variable recall of 0.750 ± 0.329, while classifications with over 1,000 images had a high and stable recall of 0.971 ± 0.0137. Models tasked with classifying species from untrained locations were less accurate, with DenseNet201 performing best with 68.7% top‐1 accuracy. Finally, we provide an open repository where ecologists can insert their image data to train and test custom species detection models for their desired ecological domain.     

Comparison of size-structured and species-level trophic networks reveals antagonistic effects of temperature on vertical trophic diversity at the population and species level

It is predicted that warmer conditions should lead to a loss of trophic levels, as larger bodied consumers, which occupy higher trophic levels, experience higher metabolic costs at high temperature. Yet, it is unclear whether this prediction is consistent with the effect of warming on the trophic structure of natural systems. Furthermore, effects of temperature at the species level, which arise through a change in species composition, may differ from those at the population level, which arise through a change in population structure. We investigate this by building species-level trophic networks, and size-structured trophic networks, as a proxy for population structure, for 18 648 stream fish communities, from 4 145 234 individual fish samples, across 7024 stream locations in France from 1980 to 2008. We estimated effects of temperature on total trophic diversity (total number of nodes), vertical trophic diversity (mean and maximum trophic level) and distribution of biomass across trophic level (correlation between trophic level and biomass) in these networks. We found a positive effect of temperature on total trophic diversity in both species- and size-structured trophic networks. We found that maximum trophic level and biomass distribution decreased in species-level and size-structured trophic networks, but the mean trophic level decreased only in size-structured trophic networks. These results show that warmer temperatures associate with a lower vertical trophic diversity in size-structured networks, and a higher one in species-level networks. This suggests that vertical trophic diversity is shaped by antagonistic effects of temperature on population structure and on species composition. Our results hence demonstrate that effects of temperature do not only differ across trophic levels, but also across levels of biological organisation, from population to species level, implying complex changes in network structure and functioning with warming.