PLOS Computational Biology

Spontaneous network bursts, the intervals of intense network-wide activity interleaved with longer periods of sparse activity, are a hallmark phenomenon observed in cortical networks at postnatal developmental stages. Generation, propagation and termination of network bursts depend on a combination of synaptic, cellular and network mechanisms; however, the interplay between these mechanisms is not fully understood. We study this interplay in silico, using a new data-driven framework for generating spiking neuronal networks fitted to the microelectrode array recordings. We recorded the network bursting activity from rat postnatal cortical networks under several pharmacological conditions. In each condition, the function of specific excitatory and inhibitory synaptic receptors was reduced in order to examine their impact on global network dynamics. The obtained data was used to develop two complementary model fitting protocols for automatic model generation. These protocols allowed us to disentangle systematically the modeled cellular and synaptic mechanisms that affect the observed network bursts. We confirmed that the change in excitatory and inhibitory synaptic transmission in silico, consistent with pharmacological conditions, can account for the changes in network bursts relative to the control data. Reproducing the exact recorded network bursts statistics required adapting both the synaptic transmission and the cellular excitability separately for each pharmacological condition. Our results bring new understanding of the complex interplay between cellular, synaptic and network mechanisms supporting the burst dynamics. While here we focused on analysis of in vitro data, our approach can be applied ex vivo and in vivo given that the appropriate experimental data is available. New & NoteworthyWe studied the role of synaptic mechanisms in shaping the neural population activity by proposing a new method to combine experimental data and data-driven computational modeling based on spiking neuronal networks. We analyze a dataset recorded from postnatal rat cortical cultures in vitro under the pharmacological influence of excitatory and inhibitory synaptic receptor antagonists. Our computational model identifies neurobiological mechanisms necessary to reproduce the changes in population activity seen across pharmacological conditions.

BackgroundFlexibility and robustness of neuronal function are closely linked to degeneracy, the ability of distinct structural or parametric configurations to produce similar functional outcomes. At the cellular level, this often manifests as ion-channel degeneracy, in which multiple combinations of intrinsic conductances yield comparable electrophysiological phenotypes. MethodologyWe used a population-based, data-driven modelling framework to generate large ensembles of biophysically detailed CA1 pyramidal neuron models constrained by somatic electrophysiological features extracted from patch-clamp recordings in acute slices from early-birth rats. 10 reconstructed morphologies were incorporated, and model populations were analyzed using parameter correlation analysis, principal component analysis, and generalization tests to assess robustness, degeneracy, and morphology dependence of intrinsic properties. ConclusionsAcross the model population, similar somatic firing behaviours emerged from widely different combinations of intrinsic parameters, demonstrating robust two-level ion channel degeneracy both within and across morphologies. Each morphology occupied a distinct region of parameter space, indicating morphology-specific compensatory effects, while weak pairwise parameter correlations suggested distributed compensation rather than tight parameter dependencies. Even with a fixed morphology, multiple parameter subspaces supported comparable electrophysiological phenotypes. Generalization across morphologies was structure-dependent and non-reciprocal, with successful parameter similarity occurring preferentially between structurally similar neurons. Interestingly, to accurately simulate spike-frequency adaptation, it was important to retain some kinetic properties of the ion channel models as free parameters during optimization. Together, these findings show that dendrite morphology shapes the valid parameter space, and similar electrophysiology of CA1 pyramidal neurons arises from the interplay between structural variability and ion-channel diversity. This work highlights the importance of population-based modelling for capturing biological variability and provides insights into how neuronal robustness might be maintained despite substantial heterogeneity, and offers a scalable pipeline for generating biophysically realistic CA1 neuron populations for use in network simulations. Author summaryNeurons must reliably process information even though their internal components, such as ion channels and cellular shape, can vary widely from cell to cell. How stable behaviour emerges from such variability is a fundamental question in neuroscience. In this study, we explored this problem using detailed computer models of early-birth rat hippocampal CA1 pyramidal neurons, a cell type that plays a central role in learning and memory. Instead of building a single "average" neuron model, we created large populations of models that all reproduced key experimental recordings but differed in their internal parameters. We found that neurons with different shapes and different combinations of ion channels could nevertheless generate similar electrical activity. This phenomenon, known as ion channel degeneracy, allows neurons to remain functional despite biological variability or perturbations. Our results show that neuronal shape strongly influences which parameter combinations are viable, but that multiple solutions exist even for the same morphology. The population of models we provide offers a resource for future studies of early-birth CA1 pyramidal cell function and dysfunction.

It is well known that sparse coding models trained on natural images learn basis functions whose shapes resemble the receptive fields (RFs) of simple cells in the primary visual cortex (V1). However, few studies have considered how these basis functions develop during training. In particular, it is unclear whether certain types of basis functions emerge more quickly than others, or whether they develop simultaneously. In this work, we train an overcomplete sparse coding model (Sparsenet) on natural images and find that there is indeed order in the development of its basis functions, with basis functions tuned to lower spatial frequencies emerging earlier and higher spatial frequency basis functions emerging later. We observe the same trend in a biologically plausible sparse coding model (SAILnet) that uses leaky integrate-and-fire neurons and synaptically local learning rules, suggesting that this result is a general feature of sparse coding. Our results are consistent with recent experimental evidence that the distribution of optimal stimuli for driving neurons to fire shifts towards higher frequencies during normal development in mouse V1. Our analysis of sparse coding models during training yields an experimentally testable prediction for V1 development that this shift may be due in part to higher spatial frequency RFs emerging later, as opposed to a global shift towards higher frequencies across all RFs, which may also play a role. We also find that at least two explanations could account for the order of RF development: 1) high frequency RFs require more information to be specified accurately, and thus may require more visual experience in order to learn, and 2) early development of low frequency RFs improves the sparseness and fidelity of the visual representation more than early development of high frequency RFs. Author summaryWe are interested in how visual neurons learn representations of the natural world. In particular, we want to know whether certain visual features are learned by the visual cortex earlier in development than others. To address this question, we turn to a class of algorithms that can learn to represent natural scenes in a sparse fashion, with only a few neurons active at any given time (population sparseness). While sparse coding has been used extensively to model the response properties of neurons in the visual cortex, we use it here to arrive at a quantitative description of the way neurons might learn to encode visual information during development. We find that receptive fields (RFs) tuned to lower spatial frequencies develop earlier in our sparse coding models compared to high frequency RFs. If our prediction is accurate, such a description would provide a general framework for understanding the development of the functional properties of V1 neurons and serve as a guide for future experimental studies. It could also lead to new computational models that learn from input statistics, as well as advances in the design of devices that can augment or replace human vision.

Biophysically detailed neuron models are often built as a one-way pipeline in which voltage-clamp data are reduced to a single set of best-fit channel parameters, which are then combined into a deterministic spiking model. This practice discards experimentally observed variability and obscures the mechanisms by which robustness and degeneracy arise in excitable systems. Here, we reintroduce parameter variability into the Hodgkin-Huxley model and embed uncertainty and global sensitivity analysis into model construction. We digitized sodium and potassium rate constant data from the original Hodgkin and Huxley figures and used bootstrap resampling to estimate the variability of the voltage-dependent kinetic parameters. We then propagated these uncertainty estimates through a spatially extended squid axon cable model using large-scale Monte Carlo simulations. At the channel level, first-order Sobol sensitivity indices revealed that all kinetic parameters contribute to output variance in a strongly time-dependent manner, with distinct parameters controlling transient and steady-state behavior for potassium and sodium conductances. At the level of neuronal excitability, sampling hundreds of thousands of parameter sets produced a heterogeneous population of firing behaviors, including non-firing, phasic, regular, and spontaneous activity. Across stimulus amplitudes, the dominant firing mode was a single spike at stimulus onset. At the same time, the regularly firing subpopulation exhibited a broad distribution of firing rates, with a mean that matched the classic Hodgkin-Huxley prediction. In the phasic subpopulation, action potential propagation and conduction velocity varied widely yet remained consistent with experimental ranges. Finally, global sensitivity analysis during spiking shows uniformly small first-order indices but large total-order indices, indicating that excitability is primarily governed by strong interactions among parameters rather than by any single conductance or kinetic parameter. These results support a population-based view of conductance-based modeling in which biologically relevant behavior emerges from structured regions of parameter space. Author summaryNeurons are often modelled by first fitting ion-channel data to a few parameters, then integrating them into a single-neuron model. This typical method masks the fact that actual experiments show variability and that many different parameter sets can yield similar electrical behavior. In this research, we explored what happens when we keep rather than average out that variability. We revisited Hodgkin and Huxleys classic squid giant axon studies and derived ranges for sodium and potassium channel parameters by resampling digitized points from the original Hodgkin-Huxley figures using bootstrap resampling. We then ran simulations of hundreds of thousands of squid axon models, each with a unique, experimentally grounded parameter set, and analyzed the collective results. This showed that the most common response was a single action potential rather than repetitive firing, aligning with the axons role in the rapid escape response. Additionally, we discovered that no single parameter alone controls spiking; rather, it depends on interactions among multiple parameters. Our findings advocate a practical change in biophysical modeling: instead of hunting for a single best-fit model, researchers should estimate parameter uncertainty directly from data, create large ensembles that sample this uncertainty, and perform sensitivity analyses on these ensembles before choosing any model for further study.

Many insects use memories of their visual environment to adaptively drive spatial behaviours. In ants, visual memories are fundamental for navigation, whereby foragers follow long visually guided routes to foraging sites and return to the location of their nest. Whilst we understand the basic visual pathway to the memory centres (Optic Lobes to Mushroom Bodies) involved in the storage of visual information, it is still largely unknown what type of representation of visual scenes underpins view-based navigation in ants. Several experimental studies have shown ants using "higher-order" visual information - that is features extracted across the whole extent of a visual scene - which raises the question as to where these features are computed. One such experimental study showed that ants can use the proportion of a shape experienced left of their visual centre to learn and recapitulate a route, a feature referred to as "fractional position of mass" (FPM). In this work, we use a simple model constrained by the known neuroanatomy and information processing properties of the Mushroom Bodies to explore whether the use of the FPM could be a resulting factor of the bilateral organisation of the insect brain, all the whilst assuming a "retinotopic" view representation. We demonstrate that such bilaterally organised memory models can implicitly encode the FPM learned during training. We find that balancing the "quality" of the memory match across left and right hemispheres allows a trained model to retrieve the FPM defined direction, even when the model is tested with other shapes, as demonstrated by ants. The result is shown to be largely independent of model parameter values, therefore suggesting that some aspects of higher-order processing of a visual scene may be emergent from the structure of the neural circuits, rather than computed in discrete processing modules. Author summaryMany insects are excellent visual navigators, often relying on visual memories to follow long foraging routes and return safely to their nest location. We have a good understanding of the neural substrates supporting the storage of visual memories in ants. However, it is still largely unknown what type of representation of visual scenes underpins the functions of visual navigation. Experimental studies have shown ants using "higher-order" features as part of navigation, that is features that are extracted across the whole extent of a visual scene. Using an anatomically constrained model of the insect memory centers, we address the question of whether the use of higher-order visual features may be emergent from the overall architecture of the vision-to-memory pathways. We find that balancing the quality of left and right visual memory matches provides an explanation for some higher-order visual processing and visual cognition shown in experiments with ants. Overall, this constitutes a contribution to our understanding of visual cognition and the processing of visual scenes used in navigational tasks. We additionally postulate a novel mechanism ants may use to navigate, which is supported by the bilateral structure of the insect brain.

Drug resistance in cancer is shaped not only by evolutionary processes but also by eco-evolutionary interactions between tumor subpopulations. These interactions can support the persistence of resistant cells even in the absence of treatment, undermining standard aggressive therapies and motivating drug holiday-based approaches that leverage ecological dynamics. A key challenge in implementing such strategies is efficiently identifying interaction between drug-sensitive and drug-resistant subpopulations. Evolutionary game theory provides a framework for characterizing these interactions. We investigate whether spatial patterns in single time-point images of cell populations can reveal the underlying game theoretic interactions between sensitive and resistant cells. To achieve this goal, we develop an agent-based model in which cell reproduction is governed by local game-theoretic interactions. We compute a suite of spatial statistics on single time-point images from the agent-based model under a range of games being played between cells. We quantify the informativeness of each spatial statistic and demonstrate that a simple machine learning model can classify the type of game being played. Our findings suggest that spatial structure contains sufficient information to infer ecological interactions. This work represents a step toward clinically viable tools for identifying cell-cell interactions in tumors, supporting the development of ecologically informed cancer therapies. Author summaryDrug resistance is a major challenge in cancer treatment, often leading to relapse despite initially successful therapy. While mutations are a key driver, ecological interactions between drug-sensitive and drug-resistant cells also play a critical role. These interactions are complex and dynamic, and few molecular biomarkers exist, making them difficult to study and account for in treatment planning. We use evolutionary game theory, a framework for quantifying interactions between cells, to investigate whether it is possible to infer these interactions using just a single time-point image of the cells. We develop an agent-based model where cells reproduce based on local interactions and quantify the resulting patterns in how cells are distributed across space using a suite of spatial statistics. We find that specific interaction types produce distinct spatial patterns that are evident in these metrics, and we train a simple machine learning model to classify the interaction type based on the metrics. Our results suggest that spatial data alone can offer valuable insights into tumor dynamics, potentially enabling more informed and adaptable cancer treatments based on eco-evolutionary principles.

We present a Recurrent Vision Transformer (Recurrent ViT) that integrates a capacity-limited spatial memory module with self-attention to emulate primate-like visual attention. Trained via reinforcement learning on a spatially cued orientation-change detection task, our model exhibits hallmark behavioral signatures of primate attention--including improved detection accuracy and faster reaction times for cued stimuli that scale with cue validity. Analysis of its self-attention maps reveals rich temporal dynamics: spatial biases induced by cues are maintained during blank intervals and reactivated prior to anticipated stimulus changes, mirroring the top-down modulation observed in primate studies. Moreover, targeted manipulations of internal attention weights yield performance changes analogous to those produced by microstimulation in attentional control regions such as the frontal eye fields and superior colliculus. These findings demonstrate that embedding recurrent, memory-driven mechanisms within transformer architectures may provide a computational framework for linking artificial and biological attention

Ambiguous sensory information can lead to spontaneous alternations between perceptual states, recently shown to extend to tactile perception. The authors recently proposed a simplified form of tactile rivalry which evokes two competing percepts for a fixed difference in input amplitudes across antiphase, pulsatile stimulation of the left and right fingers. This study addresses the need for a tactile rivalry model that captures the dynamics of perceptual alternations and that incorporates the structure of the somatosensory system. The model features hierarchical processing with two stages; a first stage resolves perceptual competition, leading to perceptual alternations; and a second stage encodes perceptual interpretations. The first stage could be located downstream of brainstem nuclei and the second stage could be located within the primary somatosensory cortex (area 3b). The model captures dynamical features specific to the tactile rivalry percepts and produces general characteristics of perceptual rivalry: input strength dependence of dominance times (Levelts proposition II), short-tailed skewness of dominance time distributions and the ratio of distribution moments. The presented modelling work leads to experimentally testable predictions. The same hierarchical model could generalise to account for percept formation, competition and alternations for bistable stimuli that involve pulsatile inputs from the visual and auditory domains. Author summaryPerceptual ambiguity involving the touch sensation has seen increased recent interest. It provides interesting opportunity to explore how our perceptual experience is resolved by dynamic computations in the brain. We recently proposed a simple form of tactile rivalry where stimuli consisted of antiphase sequences of high and low intensity pulses delivered to the right and left index fingers. The stimulus can be perceived as either one simultaneous pattern of vibration on both hands, or as a pattern of vibrations that jumps from one hand to the other, giving a sensation of apparent movement. During long presentation of the stimuli, ones perception switches every 5-20 seconds between these two interpretations, a phenomenon called tactile perceptual bistability. This study presents the first computational model for tactile bistability and is based on the structure of sensory brain areas. The model captures important characteristics of perceptual interpretations for tactile rivalry. We offer predictions in terms of how left-right tactile intensity differences are encoded and propose a location for the encoding of perceptual interpretations in sensory brain areas. The model provides a generalisable framework that can make useful predictions for future behavioural experiments with tactile and other types of stimuli.

Neural population models are widely used to interpret electroencephalography (EEG), yet their relationships remain far less systematically understood than those among single-neuron models. More fundamentally, it remains unclear whether EEG can support a uniquely plausible population-level mechanism, or whether multiple structurally distinct models can explain the data equally well. To address this question, we combine comparative analysis of canonical model families with grammar-based generation of new candidate architectures. We assembled 17 canonical neural mass and phenomenological models and embedded them in a shared structural space. From their common processes, we defined a probabilistic grammar over interpretable dynamical components and developed ENEEGMA (Exploring Neural EEG Model Architectures), a Julia-based framework for grammar-based model generation, simulation, and parameter optimization, to generate additional candidate models. We then assessed both canonical and generated models by fitting them to EEG independent-component spectra from four datasets for each condition: resting state and steady-state visual evoked potentials. Canonical models formed six structural clusters. Across conditions, compact low-dimensional polynomial oscillators performed best overall, with Montbrio-Pazo-Roxin, FitzHugh-Nagumo, and Stuart-Landau models offering the best balance of fit quality, stability, and simplicity. Grammar-based exploration further showed that the space of viable EEG node models extends beyond canonical formulations: even a restricted search over 1,000 generated models produced compact alternatives competitive with nearly all canonical families and achieving the strongest cluster-level SSVEP fits. Together, these findings suggest that EEG spectra constrain plausible neural population mechanisms without uniquely determining them. Beyond this, grammar-based model exploration provides a principled, data-driven framework for EEG-constrained model discovery. Author summaryElectroencephalography (EEG) lets us measure brain activity non-invasively, but the signals are indirect, so we rely on mathematical models to explain how neural populations generate them. Many such models exist, yet it is unclear whether standard models cover the full range of plausible explanations for EEG data, or whether several very different models can explain the same signal equally well. In this study, we compared a broad set of established neural population models and then used a grammar-based equation discovery framework to automatically generate new candidate models from interpretable building blocks. We found that simple low-dimensional oscillator models often matched EEG spectra better than more complex canonical models. We also found that newly generated models could perform nearly as well as, and sometimes better than, established ones, especially for stimulus-driven responses. These results suggest that EEG spectra alone may not be enough to identify a unique underlying neural mechanism. More broadly, our work shows how automated, biologically informed model generation can help to compare, understand, expand, and test the space of candidate neural population models.

The conception of seizures as abnormal synchronies of large neuronal populations has been confirmed by numerous electrophysiological studies, including recent imaging of travelling seizure waves across the neocortex. This traditional viewpoint has been challenged by the finding that during some seizures, neurons with high firing rates are remarkably rare and sparsely distributed into clusters. Reconciliation of these seemingly contradictory descriptions has attracted much attention, raising questions such as how (or if) macroscopic seizure waves arise from these microscope neuronal clusters, and more generally, how other features of macroscopic, clinical seizures arise from microscopic dynamics. Answers to these questions are crucial to the understanding of epilepsy, and could guide development of drugs and other interventions that act at the microscopic level to effect macroscopic improvement.\n\nRelationships between microscopic and macroscopic processes are addressed by the field of statistical physics, offering explanations for how macroscopic quantities such as pressure and temperature arise from microscopic interactions between molecules. Here we hypothesize that these methods could also provide insight between the macroscopic and microscopic dynamics of seizure behavior. We constructed a model of the neocortex composed of small domains, each representing a cluster of neurons. Models with and without refractory periods were studied. Allowing seizures to spread among the clusters in a probabilistic fashion produced a \"cellular automaton\" amenable to the methods of statistical physics. We thereby showed that the model harbors a continuous phase transition allowing possible explanations for the emergence of seizure waves from microscopic neuronal clusters, and for a surprisingly wide variety of seizure properties. Moreover, the model is easy to use because it requires only a small number of intuitively understood rules and is computationally efficient. We hope that these insights from statistical physics will contribute to the understanding of epilepsy and to the identification of new therapeutic measures.\n\nAuthor summaryEpilepsy is a common neurological disease characterized by devastating, unpredictable seizures. Extensive research is aimed at improving the treatment of epilepsy through better understanding of how seizures start and spread, but basic questions remain unanswered. Do seizures start as waves of overactive neuronal activity, or as small clusters of activity as suggested by recent data? How do clinical properties of seizures emerge from interactions between small groups of neurons? And would understanding this emergence lead to better treatment?\n\nWe address these questions with a mathematical model of seizure spread, using methods of physics designed to explain how quantities such as pressure and temperature emerge from interactions between molecules. The model produced small clusters of activity as observed in recent data, and the methods allowed us to show how these clusters react to increases in neuronal excitation to produce seizure waves and other clinical seizure behavior. The model thus provided possible answers to the questions above, based on new insights from the field of physics. If the model indeed represents a common pathway evoked by many pathological changes, it may inform the development of therapeutic measures such as antiepileptic drugs that act at the microscopic level to improve macroscopic behavior.

A growing number of studies indicate the possible involvement of astrocytes in triggering or modulating neurovascular coupling (NVC), i.e. the local dilation of blood vessels in the brain in response to neuronal activity. Astrocytes possess specialized subcellular compartments, named endfeet, that surround arterioles and capillaries, ideally positioned to mediate NVC. Various vasodilators have been shown to contribute to NVC, such as epoxyeicosatrienoic acid (EET), nitric oxide (NO), or prostaglandin E2 (PGE2), but the precise mechanisms underlying NVC and their variability remain to be fully elucidated. In particular, the involvement of astrocytes in this process is controversial. Recent translatome and proteomics data reveal that astrocytes and in particular endfeet are enriched in the proteins of the PGE2 pathway. However, how the latter could contribute to NVC remains to be characterized. Here, we develop a computational model of astrocyte-mediated NVC that recapitulates these findings and describes Ca2+ and PGE2 signaling in astrocytes, NO release by neurons, and arteriole diameter dynamics using ordinary differential equations. The model successfully reproduces the dynamics of arteriole diameter change during hyperemia from in vivo neocortical recordings in awake mice. Our simulations suggest that the astrocyte PGE2 pathway could be responsible for the late response of NVC at the arteriolar level. We further observe that PIP2-derived diacylglycerol plays a major role in driving arteriole diameter dynamics in our model, while phosphatidic acid-derived diacylglycerol, which is calcium-dependent, mainly acts as an amplifier of this response. Finally, a spatial implementation of the model using a simplified astrocyte geometry suggests that NVC is more efficient when synaptic stimulation occurs at the endfoot level rather than at other astrocytic compartments. Overall, this computational study suggests a partial role for astrocyte-mediated PGE2 release in NVC and points to astrocyte perivascular processes as sub-compartments that are ideally positioned and equipped to mediate NVC. Author summaryIn the brain, the local blood flow is regulated to meet neuronal energy demand by modulating the dilation of neighboring blood vessels. The mechanisms driving this process, known as neurovascular coupling (NVC), remain debated and are likely to differ depending on the physiological context. Recent evidence points to astrocytes, a cell type possessing specialized protrusions called "endfeet", that envelop the entire brain vascular tree. Contacts between synapses and endfeet have recently been reported, positioning the latter as ideal mediators of NVC. Here, we developed a computational model that simulates the signaling between neurons, astrocytes, and blood vessels. Our model successfully reproduces experimental recordings of blood vessels dilation in the brains of awake mice. Our simulations suggest that a specific signaling pathway in astrocytes, involving a molecule called prostaglandin E2, is a key driver of the late phase of NVC, occurring a few seconds after neuronal activity. Furthermore, our model indicates that the location of the stimulated synapses matters: signals sent to the astrocyte endfeet are particularly effective at controlling blood flow. This work helps clarify the active role of astrocytes in brain blood flow regulation, a process critical for healthy brain function.

Recurrent neural networks (RNNs) have proven to be highly successful in emulating human-like cognitive functions such as working memory. In recent years, RNNs are evolving to incorporate more biophysical realism to produce more plausible predictions on how cognitive tasks are solved in real neural circuits. However, there are major challenges in constructing and training networks with the complex and nonlinear properties of real neurons. A major component of the success of RNNs is that they share the same mathematical base as deep neural networks, permitting highly efficient optimization of model parameters using standard deep learning techniques. To do so, they use abstract representations of neurons which fail to capture the impact of cell-level biophysical and morphologic properties that may benefit network-level function. Expanding task-trained RNNs with biophysical properties such as dendrites and active ionic currents poses substantial challenges, as it moves these models away from the validated training regimes known to be highly effective for RNNs. To address this gap, we developed a biophysically detailed reservoir computing (BRC) framework with the goal of extracting mechanistic insights from biophysical neural models, and propose that these insights can be used to guide model choices that will work for specific categories of cognitive tasks. The BRC network was constructed with synaptically coupled excitatory and inhibitory cells, in which the excitatory cells include multicompartment biophysically active dendrites; motivated by empirical studies suggesting dendrites have desirable computational benefits (e.g. pattern classification and coincidence detection). We trained the BRC network to do a simplified working memory task where it had to maintain the representation of an extrinsic "cue" input. We studied the impact of extrinsic input time constants (fast AMPA vs slow NMDA) and location (dendrite vs soma) on the ability of a network to solve the task. Our results revealed that cue inputs through NMDA receptors are particularly efficient for solving the working memory task. Further, the properties of NMDA receptors are uniquely suited for cue inputs delivered at the dendrite, as networks trained with dendritic AMPA cue inputs failed to solve the task. Detailed examination of the cell and network dynamics that solve the task reveals distinct local network configurations and computing principles for the different types of extrinsic input. Overall, much like the body of mechanistic insights that have underpinned the success of training RNNs, this study lays the groundwork for applying the BRC framework to train biophysically detailed neural models to solve complex human-like cognitive tasks.

How perception of sensory stimuli emerges from brain activity is a fundamental question of neuroscience. To date, two disparate lines of research have examined this question. On one hand, human neuroimaging studies have helped us understand the large-scale brain dynamics of perception. On the other hand, work in animal models (mice, typically) has led to fundamental insight into the micro-scale neural circuits underlying perception. However, translating such fundamental insight from animal models to humans has been challenging. Here, using biophysical modeling, we show that the auditory awareness negativity (AAN), an evoked response associated with perception of target sounds in noise, can be accounted for by synaptic input to the supragranular layers of auditory cortex (AC) that is present when target sounds are heard but absent when they are missed. This additional input likely arises from cortico-cortical feedback and/or non-lemniscal thalamic projections and targets the apical dendrites of layer-V pyramidal neurons (PNs). In turn, this leads to increased local field potential activity, increased spiking activity in layer-V PNs, and the AAN. The results are consistent with current cellular models of conscious processing and help bridge the gap between the macro and micro levels of perception-related brain activity. Author SummaryTo date, our understanding of the brain basis of conscious perception has mostly been restricted to large-scale, network-level activity that can be measured non-invasively in human subjects. However, we lack understanding of how such network-level activity is supported by individual neurons and neural circuits. This is at least partially because conscious perception is difficult to study in experimental animals, where such detailed characterization of neural activity is possible. To address this gap, we used biophysical modeling to gain circuit-level insight into an auditory brain response known as the auditory awareness negativity (AAN). This response can be recorded non-invasively in humans and is associated with perceptual awareness of sounds of interest. Our model shows that the AAN likely arises from specific cortical layers and cell types. These data help bridge the gap between circuit- and network-level theories of consciousness, and could lead to new, targeted treatments for perceptual dysfunction and disorders of consciousness.

Calcium imaging allows recording from hundreds of neurons in vivo with the ability to resolve single cell activity. Evaluating and analyzing neuronal responses, while also considering all dimensions of the data set to make specific conclusions, is extremely difficult. Often, descriptive statistics are used to analyze these forms of data. These analyses, however, remove variance by averaging the responses of single neurons across recording sessions, or across combinations of neurons, to create single quantitative metrics, losing the temporal dynamics of neuronal activity, and their responses relative to each other. Dimensionally Reduction (DR) methods serve as a good foundation for these analyses because they reduce the dimensions of the data into components, while still maintaining the variance. Non-negative Matrix Factorization (NMF) is an especially promising DR analysis method for analyzing activity recorded in calcium imaging because of its mathematical constraints, which include positivity and linearity. We adapt NMF for our analyses and compare its performance to alternative dimensionality reduction methods on both artificial and in vivo data. We find that NMF is well-suited for analyzing calcium imaging recordings, accurately capturing the underlying dynamics of the data, and outperforming alternative methods in common use.

Recent studies show that astrocytic depolarization can be induced at the periphery of cortical somatosensory astrocytes, proposed to be the contact sites between neurons and astrocytes. However, specific mechanisms causing astrocytic depolarization have yet to be confirmed due to limitations in experimental techniques. Here, we constructed a computational whole-cell astrocyte model to assess which channels were responsible for astrocyte depolarization. Our simulations show that, unlike depolarization by bath application of potassium, local depolarization by potassium uptake and glutamate transporters required very large spillover and high-frequency stimulation. On the contrary, the model reproduced experimentally observed depolarizations by activating N-methyl-d-aspartate receptor (NMDAR) or -aminobutyric acid A receptor (GABAAR), on the astrocyte. Our models suggest two mechanisms for astrocyte depolarization, either by neurotransmitters or by potassium and glutamate transporters, which substantially alters the spatio-temporal dynamics of the phenomenon. These insights suggest new mechanisms of how astrocytic processes can locally regulate learning and memory.

The awake thalamus is known to be able to filter primary sensory input with and without external modulation. Through the construction and analysis of a novel computational model of a lateral geniculate thalamocortical neuron, we demonstrate how the processing of sensory retinal input is influenced by the underlying thalamic dynamic state. Our model, using only currents verified against expression data from publicly available datasets, is the first to produce five experimentally established distinct dynamic firing regimes. We demonstrate that the thalamocortical cell transitions between these dynamic states in response to glutamatergic signals from the cortex or cholinergic arousal signals coming from the brainstem. We focus on signal processing in the model dynamic states associated with the awake thalamic alpha rhythm where we find that the ability of retinal inputs to generate thalamic spikes is a balance between the timing of retinal spikes, the excitability break imposed by the M-current, and the decay time of the L-type calcium current. Finally, we explore how these two currents help the thalamus process extra-retinal rhythmic inputs, showing the model produces entrainment to slower inhibitory and excitatory rhythms, as well as detailing the importance of nesting faster frequency rhythms within slow cycles for successful thalamic transmission. Our results suggest that the awake alpha rhythm is indirectly causal by acting as a marker for the interaction of these two currents. This biophysically-constrained lateral geniculate thalamocortical cell model generates predictions regarding rhythmic dynamics under different arousal states, thalamic control of retinogeniculate transmission, and the possible impacts neurological disorders, like schizophrenia, have on thalamic processing. Variations of this model could be used to explore the functions of higher order thalamic nuclei, thereby extending its use to investigating more complex cognitive processes. Author summaryThe thalamus generates multiple distinct brain rhythms, processes primary sensory inputs, and modulates its output using feedback signals. Previous computational models of the thalamus have typically focused on a subset of these three thalamic functions without drawing relationships among them. Here we present a novel computational thalamic cell model that unites these thalamic processes. We focus on the awake alpha rhythm, a well known thalamic oscillation, and show that it is a signature of a critical working state that enables the experimentally observed thalamic filtering of retinal signals. Additionally, we find this state is optimal for processing and passing non-sensory rhythmic signals. Our model generates testable predictions about which ionic currents control the transmission of external signals. It highlights the roles of two currents from our model that do not have specified functions in the awake thalamus in previous computational models. The work concludes with hypotheses about why neurological disorders that perturb the thalamus from this alpha rhythm working state lead to significant processing errors locally within the thalamus and globally within the brain.

AbstractAlthough infectious disease dynamics are often analyzed at the macro-scale, increasing numbers of drug-resistant infections highlight the importance of within-host modeling that simultaneously solves across multiple scales to effectively respond to epidemics. We review multiscale modeling approaches for complex, interconnected biological systems and discuss critical steps involved in building, analyzing, and applying such models within the discipline of model credibility. We also present our two tools: CaliPro, for calibrating multiscale models (MSMs) to datasets, and tunable resolution, for fine- and coarse-graining sub-models while retaining insights. We include as an example our work simulating infection with Mycobacterium tuberculosis to demonstrate modeling choices and how predictions are made to generate new insights and test interventions. We discuss some of the current challenges of incorporating novel datasets, rigorously training computational biologists, and increasing the reach of MSMs. We also offer several promising future research directions of incorporating within-host dynamics into applications ranging from combinatorial treatment to epidemic response.

The inherent noise of neural systems makes it difficult to construct models which accurately capture experimental measurements of their activity. While much research has been done on how to efficiently model neural activity with descriptive models such as linear-nonlinear-models (LN), Bayesian inference for mechanistic models has received considerably less attention. One reason for this is that these models typically lead to intractable likelihoods and thus make parameter inference difficult. Here, we develop an approximate Bayesian inference scheme for a fully stochastic, biophysically inspired model of glutamate release at the ribbon synapse, a highly specialized synapse found in different sensory systems. The model translates known structural features of the ribbon synapse into a set of stochastically coupled equations. We approximate the posterior distributions by updating a parametric prior distribution via Bayesian updating rules and show that model parameters can be efficiently estimated for synthetic and experimental data from in vivo two-photon experiments in the zebrafish retina. Also, we find that the model captures complex properties of the synaptic release such as the temporal precision and outperforms a standard GLM. Our framework provides a viable path forward for linking mechanistic models of neural activity to measured data.

Integrating data from multiple experiments is common practice in systems neuroscience but it requires inter-experimental variability to be negligible compared to the biological signal of interest. This requirement is rarely fulfilled; systematic changes between experiments can drastically affect the outcome of complex analysis pipelines. Modern machine learning approaches designed to adapt models across multiple data domains offer flexible ways of removing inter-experimental variability where classical statistical methods often fail. While applications of these methods have been mostly limited to single-cell genomics, in this work, we develop a theoretical framework for domain adaptation in systems neuroscience. We implement this in an adversarial optimization scheme that removes inter-experimental variability while preserving the biological signal. We compare our method to previous approaches on a large-scale dataset of two-photon imaging recordings of retinal bipolar cell responses to visual stimuli. This dataset provides a unique benchmark as it contains biological signal from well-defined cell types that is obscured by large inter-experimental variability. In a supervised setting, we compare the generalization performance of cell type classifiers across experiments, which we validate with anatomical cell type distributions from electron microscopy data. In an unsupervised setting, we remove inter-experimental variability from data which can then be fed into arbitrary downstream analyses. In both settings, we find that our method achieves the best trade-off between removing inter-experimental variability and preserving biological signal. Thus, we offer a flexible approach to remove inter-experimental variability and integrate datasets across experiments in systems neuroscience. Code available at https://github.com/eulerlab/rave.

Human walking is intrinsically variable. For example, there is considerable stride to stride variability even when walking speed is constant. This variability is due to uncertainty in the sensorimotor system and the environment, and is shaped by both musculoskeletal dynamics (e.g. joint stiffness and damping originating from muscles) and the control strategy used to mitigate the effects of uncertainty. Yet, insight into how sensorimotor noise shapes walking variability is limited due to a lack of experimental methods to assess sensorimotor noise and control strategies during walking. Simulations that account for uncertainty can elucidate how sensorimotor noise affects movement variability but due to numerical challenges, accounting for sensorimotor noise is not common in simulations of walking. Existing simulations have hugely simplified musculoskeletal dynamics (e.g. no muscles), the control policy (e.g. pre-defined feedback loops), or sensorimotor noise sources (e.g. only motor noise). Here, we performed stochastic optimal control simulations of walking based on a model with 9 degrees of freedom and 18 muscles to study how the level of sensory and motor noise influences walking. We solved for feedforward muscle excitations and full-state time-varying feedback gains that minimised expected effort while generating periodic, and hence stable, gait patterns. To enable these simulations, we approximated the state distribution with a Gaussian and used an unscented transform to propagate the state covariance. Resulting optimisation problems were solved with direct collocation. Sensorimotor noise level had a small effect on the mean kinematics but shaped kinematic and muscle activity variability as well as expected effort. Although simulations underestimated the magnitude of experimental positional variability, they captured its structure. In agreement with experimental results, the control policy prioritised limiting variability of centre of mass kinematics and minimal swing foot clearance over limiting joint angle variability. Hence, our simulations suggest that effort minimisation underlies these observations. Author summaryWhen performing a movement multiple times, each repetition will be slightly different due to random disturbances in the neural signals used to control movement, i.e. sensorimotor noise. Because it is difficult to measure inside the nervous system of a moving person, computer simulations are used to study movement control. They found that both sensorimotor noise and musculoskeletal mechanics determine how people control arm movements and standing. However, there are no simulations of walking that systematically evaluated how sensorimotor noise level influences walking kinematics because they pose computational challenges. Here, we proposed and used an approach for minimal effort simulations of walking in the presence of uncertainty. We imposed forward speed and stability but not kinematics. We found that the level of sensorimotor noise had little effect on the mean movement but a strong effect on the variability and the expected effort. The control strategy prioritised reducing the variability of the centre of mass position and swing foot clearance over reducing the variability of individual joint angles, which is also observed in experiments. Interestingly, strict control of centre of mass position and foot clearance in our simulations emerged from minimising effort.