Biology subjects

Dora, S.

Publications and source records attributed to Dora, S..

9 recordsLinked to original sources

bioRxiv2023

Diverse and flexible behavioral strategies arise in recurrent neural networks trained on multisensory decision making

Behavioral variability across individuals leads to substantial performance differences during cognitive tasks, although its neuronal origin and mechanisms remain elusive. Here we use recurrent neural networks trained on a multisensory decision-making task to investigate inter-subject behavioral variability. By uniquely characterizing each network with a random synaptic-weights initialization, we observed a large variability in the level of accuracy, bias and decision speed across these networks, mimicking experimental observations in mice. Performance was generally improved when networks integrated multiple sensory modalities. Additionally, individual neurons developed modality-, choice- or mixed-selectivity, these preferences were different for excitatory and inhibitory neurons, and the concrete composition of each network reflected its preferred behavioral strategy: fast networks contained more choice- and mixed-selective units, while accurate networks had relatively less choice-selective units. External modulatory signals shifted the preferred behavioral strategies of networks, suggesting an explanation for the recently observed within-session strategy alternations in mice.

neuroscience↗

bioRxiv2023

Predictive coding with spiking neurons and feedforward gist signalling

Predictive coding (PC) is an influential theory in neuroscience, which suggests the existence of a cortical architecture that is constantly generating and updating predictive representations of sensory inputs. Owing to its hierarchical and generative nature, PC has inspired many computational models of perception in the literature. However, the biological plausibility of existing models has not been sufficiently explored due to their use of artificial neural network features such as a non-linear, continuous, and clock-driven function approximator as basic unit of computation. Therefore, we have developed a spiking neural network for predictive coding (SNN-PC), in which neurons communicate using event-driven and asynchronous spikes. While adopting the hierarchical structure and Hebbian learning algorithms from previous PC neural network models, SNN-PC introduces two novel features: 1) a fast feedforward sweep from the input to higher areas, which generates a spatially reduced and abstract representation of input (i.e., a neural code for the gist of a scene) and provides a neurobiological alternative to an arbitrary choice of priors; and 2) a separation of positive and negative error-computing neurons, which counters the biological implausibility of a bi-directional error neuron with a very high basal firing rate. After training with the MNIST handwritten digit dataset, SNN-PC developed hierarchical internal representations and was able to reconstruct samples it had not seen during training. SNN-PC suggests biologically plausible mechanisms by which the brain may perform perceptual inference and learning in an unsupervised manner. In addition, it may be used in neuromorphic applications that can utilize its energy-efficient, event-driven, local learning, and parallel information processing nature. Author summaryHow does the brain seamlessly perceive the world, in the midst of chaotic sensory barrage? Rather than passively relaying information that sensory organs pick up from the external world along the cortical hierarchy for a series of feature extractions, it actively gathers statistical regularities from sensory inputs to track causal relationships between physical properties of external objects and the body. In other words, the brains perceptual apparatus is constantly trying to make sense of the incoming streams of sensory input and represent the subjects current situation by building and maintaining internal models of the world and body. While this constructivist theme in understanding perception has been pervasive across multiple disciplines from philosophy to psychology to computer science, a comprehensive theory of brain function called predictive coding aims at unifying neural implementations of perception. In this study, we present a biologically plausible neural network for predictive coding that uses spiking neurons, Hebbian learning, and a feedforward visual pathway to perform perceptual inference and learning on images. Not only does the model show that predictive coding is well behaved under the biological constraint of spiking neurons, but it also provides deep learning and neuromorphic communities with novel paradigms of learning and computational architectures inspired by the natures most intelligent system, the brain.

neuroscience↗

bioRxiv2022

Vascular fungi alter the apoplastic purinergic signaling in plants by deregulating the homeostasis of extracellular ATP and its metabolite adenosine

Purinergic signaling activated by extracellular nucleotides and their derivative nucleosides trigger sophisticated signaling networks. The outcome of these pathways determine the capacity of the organism to survive under challenging conditions. Both extracellular ATP (eATP) and Adenosine (eAdo) act as secondary messengers in mammals, essential for immunosuppressive responses. Despite the clear role of eATP as a plant damage-associated molecular pattern, the function of its nucleoside, eAdo, and of the eAdo/eATP balance in plant stress response remain to be fully elucidated. This is particularly relevant in the context of plant/microbe interaction, where the intruder manipulates the extracellular matrix. Here, we identify Ado as a main elicitor secreted by the vascular fungus Fusarium oxysporum. We show that eAdo modulates the plants susceptibility to fungal colonization by altering eATP-mediated physiological immune responses, such as apoplastic pH and calcium homeostasis. Our work indicates that plant pathogens actively imbalance the eAdo/eATP levels as a virulence mechanism. One Sentence SummaryThe apoplastic Adenosine/ATP balance is a messenger for plant defense and can be manipulated by the fungal pathogen F. oxysporum.

plant biology↗

bioRxiv2021

Impairment of the cellulose degradation machinery enhances fungal virulence but limits reproductive fitness

Fungal endophytes grow in the apoplastic space, in constant contact with the plant cell wall (CW) that hinders microbe progression, while representing a source of nutrients. Although numerous fungal CW modifying proteins have been identified, their role during host colonization remains underexplored. Here we show that the root-infecting plant pathogen Fusarium oxysporum (Fo) does not require its complete arsenal of cellulases to infect the host plant. Quite the opposite, Fo mutants impaired in cellulose degradation become hypervirulent by enhancing the secretion of virulence factors. On the other hand, the reduction on cellulase activity had a severe negative effect on saprophytic growth and microconidia production during the final stages of the Fo infection cycle. These findings enhance our understanding on the function of plant CW degradation on the outcome of host-microbe interactions and reveal an unexpected role of cellulose degradation in a pathogens evolutionary success. TeaserUnexpectedly, fungi compromised in their capacity to degrade plant cellulose are hypervirulent but impaired in sporulation.

microbiology↗

bioRxiv2020

A chromosome-scale genome assembly for the Fusarium oxysporum strain Fo5176 to establish a model Arabidopsis-fungal pathosystem

Plant pathogens cause widespread yield losses in agriculture. Understanding the drivers of plant-pathogen interactions requires decoding the molecular dialog leading to either resistance or disease. However, progress in deciphering pathogenicity genes has been severely hampered by suitable model systems and incomplete fungal genome assemblies. Here, we report a significant improvement of the assembly and annotation of the genome of the Fusarium oxysporum (Fo) strain Fo5176. Fo comprises a large number of serious plant pathogens on dozens of plant species with largely unresolved pathogenicity factors. The strain Fo5176 infects Arabidopsis thaliana and, hence, constitutes a highly promising model system. We use high-coverage Pacific Biosciences Sequel long-read and Hi-C sequencing data to assemble the genome into 19 chromosomes and a total genome size of 67.98 Mb. The genome has a N50 of 4 Mb and a 99.1% complete BUSCO score. Phylogenomic analyses based on single-copy orthologs clearly place the Fo5176 strain in the Fo f sp. conglutinans clade as expected. We generated RNAseq data from culture medium and plant infections to train gene predictions and identified [~]18,000 genes including ten effector genes known from other Fo clades. We show that Fo5176 is able to infect cabbage and Brussel sprouts of the Brassica oleracea, expanding the usefulness of the Fo5176 model pathosystem. Finally, we performed large-scale comparative genomics analyses comparing the Fo5176 to 103 additional Fo genomes to define core and accessory genomic regions. In conjunction with the molecular tool sets available for A. thaliana, the Fo5176 genome and annotation provides a crucial step towards the establishment of a highly promising pathosystem.

genomics↗

bioRxiv2020

Deep predictive coding accounts for emergence of complex neural response properties along the visual cortical hierarchy

Predictive coding provides a computational paradigm for modelling perceptual processing as the construction of representations accounting for causes of sensory inputs. Here, we develop a scalable, deep predictive coding network that is trained using a Hebbian learning rule. Without a priori constraints that would force model neurons to respond like biological neurons, the model exhibits properties similar to those reported in experimental studies. We analyze low- and high-level properties such as orientation selectivity, object selectivity and sparseness of neuronal populations in the model. As reported experimentally, image selectivity increases systematically across ascending areas in the model hierarchy. A further emergent network property is that representations for different object classes become more distinguishable from lower to higher areas. Thus, deep predictive coding networks can be effectively trained using biologically plausible principles and exhibit emergent properties that have been experimentally identified along the visual cortical hierarchy. Significance StatementUnderstanding brain mechanisms of perception requires a computational approach based on neurobiological principles. Many deep learning architectures are trained by supervised learning from large sets of labeled data, whereas biological brains must learn from unlabeled sensory inputs. We developed a Predictive Coding methodology for building scalable networks that mimic deep sensory cortical hierarchies, perform inference on the causes of sensory inputs and are trained by unsupervised, Hebbian learning. The network models are well-behaved in that they faithfully reproduce visual images based on high-level, latent representations. When ascending the sensory hierarchy, we find increasing image selectivity, sparseness and generalizability for object classification. These models show how a complex neuronal phenomenology emerges from biologically plausible, deep networks for unsupervised perceptual representation.

neuroscience↗

bioRxiv2019

Pathogen-induced pH changes regulate the growth-defense balance of plants

Environmental adaptation of organisms relies on fast perception and response to external signals, which lead to developmental changes. Plant cell growth is strongly dependent on cell wall remodeling. However, little is known about cell wall-related sensing of biotic stimuli and the downstream mechanisms that coordinate growth and defense responses. We generated genetically encoded pH sensors to determine absolute pH changes across the plasma membrane in response to biotic stress. A rapid apoplastic acidification by phosphorylation-based proton pump activation was followed by an acidification of the cortical side of the plasma membrane in response to the fungus Fusarium oxysporum. The proton chemical gradient modulation immediately reduced cellulose synthesis and cell growth and, furthermore, had a direct influence on the pathogenicity of the fungus. All these effects were dependent on the COMPANION OF CELLULOSE SYNTHASE proteins that are thus at the nexus of plant growth and defense. Hence, our discoveries show a remarkable connection between plant biomass production, immunity, and pH control, and advance our ability to investigate the plant growth-defense balance.

plant biology↗

bioRxiv2018

Measurement of infection efficiency of a major wheat pathogen using time-resolved imaging of disease progress

Infection efficiency is a key epidemiological parameter that determines the proportion of pathogen spores able to infect and cause lesions once they have landed on a susceptible plant tissue. In this study, we present an improved method to measure infection efficiency of Zymoseptoria tritici using a replicated greenhouse experiment. Z. tritici is a fungal pathogen that infects wheat leaves and causes Septoria tritici blotch (STB), a major disease of wheat worldwide.\n\nWe devised an original experimental setup, where we (i) attached living wheat leaves to metal plates allowing for time-resolved imaging of disease progress in planta. Since lesions were continuously appearing, expanding and merging during the period of up to three weeks, daily measurements were necessary for accurate counting of lesions. We also (ii) used reference membranes to characterize the density and the spatial distribution of inoculated spores on leaf surfaces. In this way, we captured the relationship between the number of lesions and the number of viable spores deposited on the leaves and estimated the infection efficiency of about 4 % from the slope of this relationship.\n\nOur study provides a proof of principle for an accurate and reliable measurement of infection efficiency of Z. tritici. The method opens opportunities for determining the genetic basis of the component of quantitative resistance that suppresses infection efficiency. This knowledge would improve breeding for quantitative resistance against STB, a control measure considered more durable than deployment of major resistance genes.

epidemiology↗

bioRxiv2018

A Deep Predictive Coding Network for Learning Latent Representations

It has been argued that the brain is a prediction machine that continuously learns how to make better predictions about the stimuli received from the external environment. It builds a model of the world around us and uses this model to infer the external stimulus. Predictive coding has been proposed as a mechanism through which the brain might be able to build such a model of the external environment. However, it is not clear how predictive coding can be used to build deep neural network models of the brain while complying with the architectural constraints imposed by the brain. In this paper, we describe an algorithm to build a deep generative model using predictive coding that can be used to infer latent representations about the stimuli received from external environment. Specifically, we used predictive coding to train a deep neural network on real-world images in a unsupervised learning paradigm. To understand the capacity of the network with regards to modeling the external environment, we studied the latent representations generated by the model on images of objects that are never presented to the model during training. Despite the novel features of these objects the model is able to infer the latent representations for them. Furthermore, the reconstructions of the original images obtained from these latent representations preserve the important details of these objects.

neuroscience↗