Interrogating theoretical models of neural computation with emergent property inference.
Ontology highlight
ABSTRACT:
SUBMITTER: Bittner SR
PROVIDER: S-EPMC8321557 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:A fundamental question in embryo morphogenesis is how a complex pattern is established in seemingly uniform tissues. During vertebrate development, neural crest cells differentiate as a continuous mass of tissue along the neural tube and subsequently split into spatially distinct migratory streams to invade the rest of the embryo. How these streams are established is not well understood. Inhibitory signals surrounding the migratory streams led to the idea that position and size of streams are determined by a pre-pattern of such signals. While clear evidence for a pre-pattern in the cranial region is still lacking, all computational models of neural crest migration published so far have assumed a pre-pattern of negative signals that channel the neural crest into streams. Here we test the hypothesis that instead of following a pre-existing pattern, the cranial neural crest creates their own migratory pathway by interacting with the surrounding tissue. By combining theoretical modeling with experimentation, we show that streams emerge from the interaction of the hindbrain neural crest and the neighboring epibranchial placodal tissues, without the need for a pre-existing guidance cue. Our model suggests that the initial collective neural crest invasion is based on short-range repulsion and asymmetric attraction between neighboring tissues. The model provides a coherent explanation for the formation of cranial neural crest streams in concert with previously reported findings and our new in vivo observations. Our results point to a general mechanism of inducing collective invasion patterns.
| S-EPMC6497294 | biostudies-literature
Project description:To maximize future rewards in this ever-changing world, animals must be able to discover the temporal structure of stimuli and then anticipate or act correctly at the right time. How do animals perceive, maintain, and use time intervals ranging from hundreds of milliseconds to multiseconds in working memory? How is temporal information processed concurrently with spatial information and decision making? Why are there strong neuronal temporal signals in tasks in which temporal information is not required? A systematic understanding of the underlying neural mechanisms is still lacking. Here, we addressed these problems using supervised training of recurrent neural network models. We revealed that neural networks perceive elapsed time through state evolution along stereotypical trajectory, maintain time intervals in working memory in the monotonic increase or decrease of the firing rates of interval-tuned neurons, and compare or produce time intervals by scaling state evolution speed. Temporal and nontemporal information is coded in subspaces orthogonal with each other, and the state trajectories with time at different nontemporal information are quasiparallel and isomorphic. Such coding geometry facilitates the decoding generalizability of temporal and nontemporal information across each other. The network structure exhibits multiple feedforward sequences that mutually excite or inhibit depending on whether their preferences of nontemporal information are similar or not. We identified four factors that facilitate strong temporal signals in nontiming tasks, including the anticipation of coming events. Our work discloses fundamental computational principles of temporal processing, and it is supported by and gives predictions to a number of experimental phenomena.
| S-EPMC7229760 | biostudies-literature
Project description:Flexible behaviors are organized by complex neural networks in the prefrontal cortex. Recent studies have suggested that such networks exhibit multiple dynamical states, and can switch rapidly from one state to another. In many complex systems such as the brain, the early-warning signals that may predict whether a critical threshold for state transitions is approaching are extremely difficult to detect. We hypothesized that increases in firing irregularity are a crucial measure for predicting state transitions in the underlying neuronal circuits of the prefrontal cortex. We used both experimental and theoretical approaches to test this hypothesis. Experimentally, we analyzed activities of neurons in the prefrontal cortex while monkeys performed a maze task that required them to perform actions to reach a goal. We observed increased firing irregularity before the activity changed to encode goal-to-action information. Theoretically, we constructed theoretical generic neural networks and demonstrated that changes in neuronal gain on functional connectivity resulted in a loss of stability and an altered state of the networks, accompanied by increased firing irregularity. These results suggest that assessing the temporal pattern of neuronal fluctuations provides important clues regarding the state stability of the prefrontal network. We also introduce a novel scheme that the prefrontal cortex functions in a metastable state near the critical point of bifurcation. According to this scheme, firing irregularity in the prefrontal cortex indicates that the system is about to change its state and the flow of information in a flexible manner, which is essential for executive functions. This metastable and/or critical dynamical state of the prefrontal cortex may account for distractibility and loss of flexibility in the prefrontal cortex in major mental illnesses such as schizophrenia.
| S-EPMC3857743 | biostudies-literature
Project description:We propose that synapses may be the workhorse of the neuronal computations that underlie probabilistic reasoning. We built a neural circuit model for probabilistic inference in which information provided by different sensory cues must be integrated and the predictive powers of individual cues about an outcome are deduced through experience. We found that bounded synapses naturally compute, through reward-dependent plasticity, the posterior probability that a choice alternative is correct given that a cue is presented. Furthermore, a decision circuit endowed with such synapses makes choices on the basis of the summed log posterior odds and performs near-optimal cue combination. The model was validated by reproducing salient observations of, and provides insights into, a monkey experiment using a categorization task. Our model thus suggests a biophysical instantiation of the Bayesian decision rule, while predicting important deviations from it similar to the 'base-rate neglect' observed in human studies when alternatives have unequal prior probabilities.
| S-EPMC2921378 | biostudies-literature
Project description:Current theoretical models of the commons assert that common-pool resources can only be managed sustainably with clearly defined boundaries around both communities and the resources that they use. In these theoretical models, open access inevitably leads to a tragedy of the commons. However, in many open-access systems, use of common-pool resources seems to be sustainable over the long term (i.e., current resource use does not threaten use of common-pool resources for future generations). Here, we outline the conditions that support sustainable resource use in open property regimes. We use the conceptual framework of complex adaptive systems to explain how processes within and couplings between human and natural systems can lead to the emergence of efficient, equitable, and sustainable resource use. We illustrate these dynamics in eight case studies of different social-ecological systems, including mobile pastoralism, marine and freshwater fisheries, swidden agriculture, and desert foraging. Our theoretical framework identifies eight conditions that are critical for the emergence of sustainable use of common-pool resources in open property regimes. In addition, we explain how changes in boundary conditions may push open property regimes to either common property regimes or a tragedy of the commons. Our theoretical model of emergent sustainability helps us to understand the diversity and dynamics of property regimes across a wide range of social-ecological systems and explains the enigma of open access without a tragedy. We recommend that policy interventions in such self-organizing systems should focus on managing the conditions that are critical for the emergence and persistence of sustainability.
| S-EPMC6304945 | biostudies-literature
Project description:We have characterized and quantified a form of bacterial chemotaxis that manifests only as an emergent property by measuring symmetry breaking in a swarm of Myxococcus xanthus exposed to a two-dimensional nutrient gradient from within an agar substrate. M. xanthus chemotaxis requires cell-cell contact and coordinated motility, as individual motile cells exhibit only nonvectorial movement in the presence of a nutrient gradient. Genes that specifically affect M. xanthus chemotaxis include at least 10 of the 53 that express enhancer binding proteins of the NtrC-like class, an indication that this behavior is controlled through transcription, most likely by a complex signal transduction network.
| S-EPMC2566203 | biostudies-literature
Project description:Intracellular transport is an essential function in eucaryotic cells, facilitated by motor proteins-proteins converting chemical energy into kinetic energy. It is understood that motor proteins work in teams enabling unidirectional and bidirectional transport of intracellular cargo over long distances. Disruptions of the underlying transport mechanisms, often caused by mutations that alter single motor characteristics, are known to cause neurodegenerative diseases. For example, phosphorylation of kinesin motor domain at the serine residue is implicated in Huntington's disease, with a recent study of phosphorylated and phosphomimetic serine residues indicating lowered single motor stalling forces. In this article we report the effects of mutations of this nature on transport properties of cargo carried by multiple wild-type and mutant motors. Results indicate that mutants with altered stall forces might determine the average velocity and run-length even when they are outnumbered by wild type motors in the ensemble. It is shown that mutants gain a competitive advantage and lead to an increase in the expected run-length when the load on the cargo is in the vicinity of the mutant's stalling force or a multiple of its stalling force. A separate contribution of this article is the development of a semi-analytic method to analyze transport of cargo by multiple motors of multiple types. The technique determines transition rates between various relative configurations of motors carrying the cargo using the transition rates between various absolute configurations. This enables a computation of biologically relevant quantities like average velocity and run-length without resorting to Monte Carlo simulations. It can also be used to introduce alterations of various single motor parameters to model a mutation and to deduce effects of such alterations on the transport of a common cargo by multiple motors. Our method is easily implementable and we provide a software package for general use.
| S-EPMC5094777 | biostudies-literature
Project description:Recent technological advances may lead to the development of small scale quantum computers capable of solving problems that cannot be tackled with classical computers. A limited number of algorithms has been proposed and their relevance to real world problems is a subject of active investigation. Analysis of many-body quantum system is particularly challenging for classical computers due to the exponential scaling of Hilbert space dimension with the number of particles. Hence, solving problems relevant to chemistry and condensed matter physics are expected to be the first successful applications of quantum computers. In this paper, we propose another class of problems from the quantum realm that can be solved efficiently on quantum computers: model inference for nuclear magnetic resonance (NMR) spectroscopy, which is important for biological and medical research. Our results are based on three interconnected studies. Firstly, we use methods from classical machine learning to analyze a dataset of NMR spectra of small molecules. We perform a stochastic neighborhood embedding and identify clusters of spectra, and demonstrate that these clusters are correlated with the covalent structure of the molecules. Secondly, we propose a simple and efficient method, aided by a quantum simulator, to extract the NMR spectrum of any hypothetical molecule described by a parametric Heisenberg model. Thirdly, we propose a simple variational Bayesian inference procedure for extracting Hamiltonian parameters of experimentally relevant NMR spectra.
| S-EPMC7643990 | biostudies-literature
Project description:The skyline plot is a graphical representation of historical effective population sizes as a function of time. Past population sizes for these plots are estimated from genetic data, without a priori assumptions on the mathematical function defining the shape of the demographic trajectory. Because of this flexibility in shape, skyline plots can, in principle, provide realistic descriptions of the complex demographic scenarios that occur in natural populations. Currently, demographic estimates needed for skyline plots are estimated using coalescent samplers or a composite likelihood approach. Here, we provide a way to estimate historical effective population sizes using an Approximate Bayesian Computation (ABC) framework. We assess its performance using simulated and actual microsatellite datasets. Our method correctly retrieves the signal of contracting, constant and expanding populations, although the graphical shape of the plot is not always an accurate representation of the true demographic trajectory, particularly for recent changes in size and contracting populations. Because of the flexibility of ABC, similar approaches can be extended to other types of data, to multiple populations, or to other parameters that can change through time, such as the migration rate.
| S-EPMC5518730 | biostudies-literature
Project description:Noise-induced heterogeneity in gene expression is an inherent reality for cells. However, it is not well understood how noise strength changes for a single gene while the host cell is aging. Using a state-of-the-art microfluidic platform, we measure noise dynamics in aging yeast cells by tracking the generation-specific activity of the canonical GAL1 promoter. We observe noise reduction during normal aging of a cell, followed by a short catastrophe phase in which noise increased. We hypothesize that aging-associated increases in chromatin state transitions are behind the observed noise reduction and a stochastic model provides quantitative support to the proposed mechanism. Noise trends measured from strains with altered GAL1 promoter dynamics (constitutively active, synthetic with nucleosome-disfavoring sequences, and in the absence of RPD3, a global remodeling regulator) lend further support to our hypothesis. Observing similar noise dynamics from a different promoter (HHF2) provides support to the generality of our findings.Gene expression is a noisy process, but it is not known how noise in gene expression changes during the aging of single cells. Here the authors show that noise decreases during normal aging, and provide support for aging-associated increases in chromatin state transitions governing noise reduction.
| S-EPMC5613028 | biostudies-literature