Modular-integrative modeling: a new framework for building brain models that blend biological realism and functional performance.

This Perspective presents the Modular-Integrative Modeling approach, a novel framework in neuroscience for developing brain models that blend biological realism with functional performance to provide a holistic view on brain function in interaction with the body and environment.

Contextual encoder-decoder network for visual saliency prediction.

Predicting salient regions in natural images requires the detection of objects that are present in a scene. To develop robust representations for this challenging task, high-level visual features at multiple spatial scales must be extracted and augmented with contextual information. However, existing models aimed at explaining human fixation maps do not incorporate such a mechanism explicitly. Here we propose an approach based on a convolutional neural network pre-trained on a large-scale image classification task. The architecture forms an encoder-decoder structure and includes a module with multiple convolutional layers at different dilation rates to capture multi-scale features in parallel. Moreover, we combine the resulting representations with global scene information for accurately predicting visual saliency. Our model achieves competitive and consistent results across multiple evaluation metrics on two public saliency benchmarks and we demonstrate the effectiveness of the suggested approach on five datasets and selected examples. Compared to state of the art approaches, the network is based on a lightweight image classification backbone and hence presents a suitable choice for applications with limited computational resources, such as (virtual) robotic systems, to estimate human fixations across complex natural scenes. Our TensorFlow implementation is openly available at https://github.com/alexanderkroner/saliency.

Running Large-Scale Simulations on the Neurorobotics Platform to Understand Vision - The Case of Visual Crowding.

Traditionally, human vision research has focused on specific paradigms and proposed models to explain very specific properties of visual perception. However, the complexity and scope of modern psychophysical paradigms undermine the success of this approach. For example, perception of an element strongly deteriorates when neighboring elements are presented in addition (visual crowding). As it was shown recently, the magnitude of deterioration depends not only on the directly neighboring elements but on almost all elements and their specific configuration. Hence, to fully explain human visual perception, one needs to take large parts of the visual field into account and combine all the aspects of vision that become relevant at such scale. These efforts require sophisticated and collaborative modeling. The Neurorobotics Platform (NRP) of the Human Brain Project offers a unique opportunity to connect models of all sorts of visual functions, even those developed by different research groups, into a coherently functioning system. Here, we describe how we used the NRP to connect and simulate a segmentation model, a retina model, and a saliency model to explain complex results about visual perception. The combination of models highlights the versatility of the NRP and provides novel explanations for inward-outward anisotropy in visual crowding.

A Trade-Off Between Sporangia Size and Number Exists in the Potato Late Blight Pathogen Phytophthora infestans, and Is Not Altered by Biotic and Abiotic Factors.

The negative relationship between offspring size and number is a classic example of trade-off between life-history traits, reported many times in animal and plant species. Here, we wanted to ascertain whether such a trade-off occurred in the oomycete Phytophthora infestans, and whether it was impacted by biotic and abiotic factors. We thus conducted three infection experiments under controlled conditions and measured the number and the size of sporangia (asexual propagules) produced on potato by different P. infestans isolates. In all experiments, we observed a negative relationship between sporangia size and number, demonstrating the existence of a trade-off. Moreover, although the potato host cultivar, temperature and host of origin (tomato or potato) all affected sporangia number, sporangia size or both, none of these biotic and abiotic factors did change the trade-off. Therefore, the trade-off between sporangia size and number could maintain the polyphenism for these traits in P. infestans populations, and favors the coexistence of distinct reproductive strategies within this species. Our results emphasize the relevance to focus on the relationship between offspring size and number in other fungal plant pathogens, as well as to study the impact of offspring size on fitness-linked traits (virulence and disease lesion development) in these organisms.

The coexistence of generalist and specialist clonal lineages in natural populations of the Irish Famine pathogen Phytophthora infestans explains local adaptation to potato and tomato.

Phytophthora infestans, causing late blight on Solanaceae, is a serious threat to potato and tomato crops worldwide. P. infestans populations sampled on either potato or tomato differ in genotypes and pathogenicity, suggesting niche exclusion in the field. We hypothesized that such niche separation can reflect differential host exploitation by different P. infestans genotypes. We thus compared genotypes and phenotypes in 21 isolates sampled on potato (n = 11) or tomato (n = 10). Typing at 12 microsatellite loci assigned potato isolates to the 13_A2, 6_A1 and 1_A1 lineages, and tomato isolates to the 23_A1, 2_A1 and unclassified multilocus genotypes. Cross-inoculations on potato and tomato leaflets showed that all isolates were pathogenic on both hosts. However, tomato isolates performed much better on tomato than did potato isolates, which performed better on potato than did tomato isolates, thus revealing a clear pattern of local adaptation. Potato isolates were significantly fitter on potato than on tomato, and are best described as potato specialists; tomato isolates appear to be generalists, with similar pathogenicity on both hosts. Niche separation in the field may thus result mainly from the large fitness gap on tomato between generalists and unadapted potato specialists, while the small, but significant fitness difference on potato between both types of isolates may prevent population invasion by generalists. Extreme specialization to potato seems very costly relative to performance loss on the alternative host. This study therefore shows that local adaptation and niche separation, commonly expected to involve and generate specialists, can occur with generalists.

Nicotiflorin, rutin and chlorogenic acid: phenylpropanoids involved differently in quantitative resistance of potato tubers to biotrophic and necrotrophic pathogens.

Physiological and molecular mechanisms underlying quantitative resistance of plants to pathogens are still poorly understood, but could depend upon differences in the intensity or timing of general defense responses. This may be the case for the biosynthesis of phenolics which are known to increase after elicitation by pathogens. We thus tested the hypothesis that differences in quantitative resistance were related to differential induction of phenolics by pathogen-derived elicitors. Five potato cultivars (Solanum tuberosum, L.) spanning a range of quantitative resistance were treated with a concentrated culture filtrate (CCF) of Phytophthora infestans or purified lipopolysaccharides (LPS) from Pectobacterium atrosepticum. The kinetic of phenolics accumulation was followed and a set of typical phenolics was identified: chlorogenic acid, phenolamides and flavonols including rutin (quercetin-3-O-rutinoside) and nicotiflorin (kaempferol-3-O-rutinoside). Our results showed that CCF but not LPS induced differential accumulation of major phenolics among cultivars. Total phenolics were related with resistance to P. atrosepticum but not to P. infestans. However, nicotiflorin was inversely related with resistance to both pathogens. Rutin, but not nicotiflorin, inhibited pathogen growth in vitro at physiological concentrations. These data therefore suggest that (i) several phenolics are candidate markers for quantitative resistance in potato, (ii) some of these are pathogen specific although they are produced by a general defense pathway, (iii) resistance marker molecules do not necessarily have antimicrobial activity, and (iv) the final content of these target molecules-either constitutive or induced-is a better predictor of resistance than their inducibility by pathogen elicitors.

Quantitative resistance of potato to Pectobacterium atrosepticum and Phytophthora infestans: integrating PAMP-triggered response and pathogen growth.

While the mechanisms underlying quantitative resistance of plants to pathogens are still not fully elucidated, the Pathogen-Associated Molecular Patterns (PAMPs)-triggered response model suggests that such resistance depends on a dynamic interplay between the plant and the pathogen. In this model, the pathogens themselves or elicitors they produce would induce general defense pathways, which in turn limit pathogen growth and host colonisation. It therefore suggests that quantitative resistance is directly linked to a common set of general host defense mechanisms, but experimental evidence is still inconclusive. We tested the PAMP-triggered model using two pathogens (Pectobacterium atrosepticum and Phytophthora infestans) differing by their infectious processes and five potato cultivars spanning a range of resistance levels to each pathogen. Phenylalanine ammonia-lyase (PAL) activity, used as a defense marker, and accumulation of phenolics were measured in tuber slices challenged with lipopolysaccharides from P. atrosepticum or a concentrated culture filtrate from P. infestans. PAL activity increased following treatment with the filtrate but not with lipopolysaccharides, and varied among cultivars. It was positively related to tuber resistance to P. atrosepticum, but negatively related to tuber resistance to P. infestans. It was also positively related to the accumulation of total phenolics. Chlorogenic acid, the main phenolic accumulated, inhibited growth of both pathogens in vitro, showing that PAL induction caused active defense against each of them. Tuber slices in which PAL activity had been induced before inoculation showed increased resistance to P. atrosepticum, but not to P. infestans. Our results show that inducing a general defense mechanism does not necessarily result in quantitative resistance. As such, they invalidate the hypothesis that the PAMP-triggered model alone can explain quantitative resistance. We thus designed a more complex model integrating physiological host response and a key pathogen life history trait, pathogen growth, to explain the differences between the two pathosystems.