シンポジウム
50 理論と定量実験アプローチの融合 -シンプルな神経回路の場合-
50 Theory meets quantitative experiments -in the case of simple circuits-
座長：塚田 祐基（名古屋大学大学院 理学研究科）・Pan Chun-Liang（The Institute of Molecular Medicine, National Taiwan University Schoolof Medicine） |
|---|
| 2022年7月3日 14:03～14:23 沖縄コンベンションセンター 会議場B3・4 第6会場
4S06a-01
Theory, reimagined
*Greg J Stephens(1)
1. OIST Graduate University and Vrije Universiteit Amsterdam
Keyword: state space, dynamical systems, C. elegans, Transition matrix
Physics offers countless examples for which theoretical predictions are astonishingly powerful, such as the first detection of gravitational waves using atomic scale deformations using kilometer-scale interferometers. But it is hard to imagine a similar precision in living systems where the number and interdependencies between components simply prohibits a first-principles approach. Look no further than the challenge of the billions of neurons and trillions of connections within our own brains. In such settings how do we even identify important theoretical questions? Here we describe a systems-scale perspective in which we integrate information theory, dynamical systems and statistical physics to extract understanding directly from measurements. We demonstrate our approach with a reconstructed state space of the behavior of the nematode worm C. elegans, revealing a low-dimensional chaotic attractor with symmetric Lyapunov spectrum. We then outline a maximally predictive coarse-graining in which nonlinearities are subsumed into a linear, ensemble evolution to obtain a simple yet accurate model on multiple scales. With this coarse-graining we identify long timescales and collective states in the Langevin dynamics of a double-well potential, the Lorenz system and in worm behavior. Our "inverse" perspective provide an emergent, quantitative framework in which to seek rather than impose effective organizing principles of complex systems. | | 2022年7月3日 14:23～14:43 沖縄コンベンションセンター 会議場B3・4 第6会場
4S06a-02
探索行動を実現する感覚刺激に対する非即時的な応答
Non-instantaneous response to sensory stimuli during navigational behavior
*塚田 祐基(1)、森 郁恵(1)
1. 名古屋大学大学院
*Yuki Tsukada Tsukada(1), Ikue Mori(1)
1. Nagoya University
Keyword: navigation, information theory, quantitative experiments, non-instantaneous response
Animals sense environmental signals to generate an effective and efficient behavior for survival. Behavioral responses for sensory stimuli are often optimized as a sequence rather than the direct instantaneous reaction to a stimulus. For example, a certain extent of persistency during avoidance behavior contributes to the success of avoidance. The mechanisms that underlie the generation of non-instantaneous behavioral response to the given stimuli remains elusive. Here, we target the thermotaxis behavior of Caenorhabditis elegans as a navigational behavior and elucidate how C. elegans generates the behavioral control signals based on the sensed thermal stimuli. Our tracking system monitors calcium signals in the AFD thermosensory neuron by using fluorescence imaging during thermotaxis behavior. The time course of AFD neural activity showed discrete responses for thermal increase even with the slight continuous thermal change on a shallow thermal gradient. These discrete activities correlate directional turnings of animals during thermotaxis but not as an instantaneous response. Accumulating evidence suggests that the AIY interneuron, the direct downstream of AFD, plays crucial roles both in thermotaxis behavior and behavioral components such as reversal initiation. Simultaneous monitoring of AFD and AIY activity at the thermal increase showed that AIY activities consist of both AFD dependent and independent stochastic pulses. Single synaptic perturbation by optogenetic inhibition and laser surgery experiments suggest that the direct connection between AFD and AIY is responsible for the relationship between AFD and AIY activities. To consider the meanings of the pulse signals in AIY, we are considering a mathematical modeling approach based on “infotaxis” which utilizes information theory. We are making a model consisting of AFD evoked pulse signals and stochastic pulse signals that are attributed to the circuit status derived from non-AFD neurons. Considering the previous reports about the specific neural ablation and genetical perturbations, we propose a mechanism of navigation behavior based on the discrete sensory stimuli that enable to perform a robust and effective exploration. Our model proposes a basis for quantifying navigational behavior based on quantitative experimental data. | | 2022年7月3日 14:43～15:03 沖縄コンベンションセンター 会議場B3・4 第6会場
4S06a-03
Direct measurement of whole-brain functional connectivity in C. elegans
*Andrew Leifer(1,2), Francesco Randi(1), Anuj Sharma(1), Sophie Dvali(1)
1. Department of Physics, Princeton University, Princeton, New Jersey, USA, 2. Princeton Neuroscience Institute, Princeton University, Princeton, New Jersey, USA
Keyword: Functional connectivity, C. elegans, Neural circuits, Neural dynamics
Mathematical models of the brain are most powerful when well constrained by measurement. The nematode C. elegans is an attractive target for modeling because of its small 302 neuron nervous system and its known atlas of neural wiring and gene expression. Even in this simple worm, models to accurately predict detailed neural dynamics have so far remained elusive. One challenge has been a lack of measurements of functional connectivity to constrain models. How signals transform from one neuron to the next is unknown for many neural connections because it relies on molecular details hidden from view. To fill this gap, we systematically measure detailed properties of the transfer functions of all neural connections in the head of the nematode C. elegans, what we call functional connectivity.
We systematically perturb individual neurons in C. elegans and measure the response of all neurons in its brain to directly measure functional connectivity at brain scale and cellular resolution. We measure the sign, strength, causal direction, and temporal properties of the connections between all neuron pairs. We present a preliminary functional connectome characterizing responses to activation of neurons from over 90 of the 118 total neuron classes, aggregated across 29 animals and comprising more than 1,900 stimulation events. Our worms express a calcium indicator and a purple-light sensitive optogenetic actuator in every neuron, as well as NeuroPAL a deterministic brainbow used to identify neurons.
We investigate the stereotypy and timescale of evoked responses and compare connections to the known connectome and gene expression data. We validate our measurements against well-characterized connections, including the AFD AIY connection. And we explore in detail the role of RID which outputs neural signals almost exclusively via extrasynaptic signaling. We identify 20 neurons that respond to RID activation at timescales indicative of extrasynaptic communication. 18 of these express receptors for transmitters released by RID, and two more are consistent with receiving indirect signals from RID. This demonstrates that our direct measures of functional connectivity resolve wireless signaling between neurons consistent with gene expression data that would not otherwise be visible in the wired connectome.
The functional connectivity we measure will constrain future modeling efforts and will inform studies of structure and function in the nervous system. | | 2022年7月3日 15:03～15:23 沖縄コンベンションセンター 会議場B3・4 第6会場
4S06a-04
嗅覚系神経回路の進化的モデル選択とその制約の理論的考察
Analytical insights on the evolutionary model selection of olfactory circuits and its potential constraints
*平谷 直輝(1)
1. ハーバード大学
*Naoki Hiratani(1)
1. Harvard University, Cambridge, USA
Keyword: olfaction, neural circuit, model selection, statistical learning theory
Across species, neural circuits show remarkable regularity, suggesting that their structure has been driven by underlying optimality principles. In this talk, I discuss whether we can predict the neural circuitry of diverse species by optimizing the neural architecture to make learning as efficient as possible. I focus on the olfactory system, primarily because it has a relatively simple evolutionarily conserved structure, and because its input and intermediate layer sizes exhibit a tight allometric scaling. In mammals, it has been shown that the number of neurons in layer 2 of piriform cortex scales as the number of glomeruli (the input units) to the 3/2 power; in invertebrates, I show that the number of mushroom body Kenyon cells scales as the number of glomeruli to the 7/2 power. To understand these scaling laws, I model the olfactory system as a three layered nonlinear neural network, and analytically optimize the intermediate layer size for efficient learning from a limited number of samples. I find that a scaling law emerges robustly, both in full batch optimization and under stochastic gradient learning, but its exponent depends strongly on the number of samples, and thus on longevity. The 3/2 scaling seen in mammals is consistent with observed longevity, but the 7/2 scaling in invertebrates is not. However, when a fraction of the olfactory circuit is genetically specified, not learned, scaling becomes steeper for species with a small number of glomeruli as observed among insects. This study thus provides analytic insight into the principles underlying both allometric scaling across species and optimal architectures in artificial networks. | | 2022年7月3日 15:23～15:43 沖縄コンベンションセンター 会議場B3・4 第6会場
4S06a-05
Reading out responses of large neural population with minimal information loss
*Tatyana O. Sharpee(1)
1. Salk Institute for Biological Studies
Keyword: mutual information, population vector, sufficient statistics
Classic studies show that in many species – from leech and cricket to primate – responses of neural populations can be quite successfully read out using a measure neural population activity termed the population vector. However, despite its successes, detailed analyses have shown that the standard population vector discards substantial amounts of information contained in the responses of a neural population, and so is unlikely to accurately describe how signal communication between parts of the nervous system. I will describe recent theoretical results showing how to modify the population vector expression in order to read out neural responses without information loss. Compared to the standard population vector, the information-preserving read out includes just one additional weighting factor that describes the sharpness of neuronal nonlinearity and represents a measure of neuronal variability. In this way, more reliable neurons are weighted more than weakly tuned neurons. It is noteworthy that there is no simple expression for the information-preserving read-out when written in terms of parameters of neural tuning curves. Although noise correlation affect the amount of the information contained in the responses of the neural population, the same read-out expression continues to work when noise correlations increase or decrease in strength. These results demonstrate how to quantify information transmitted by neurons with irregular tuning curves. I will describe three approximations that make it possible to quantify information transmitted by large neural populations containing thousands of neurons. | | | |
|
|
