Single Cortical Neurons as Deep Artificial Neural Networks, bioRxiv, 2019-04-19
AbstractWe propose a novel approach based on modern deep artificial neural networks (DNNs) for understanding how the morpho-electrical complexity of neurons shapes their inputoutput (IO) properties at the millisecond resolution in response to massive synaptic input. The IO of integrate and fire point neuron is accurately captured by a DNN with a single unit and one hidden layer. A fully connected DNN with one hidden layer faithfully replicated the IO relationship of a detailed model of Layer 5 cortical pyramidal cell (L5PC) receiving AMPA and GABAA synapses. However, when adding voltage-gated NMDA-conductances, a temporally-convolutional DNN with seven layers was required. Analysis of the DNN filters provides new insights into dendritic processing shaping the IO properties of neurons. This work proposes a systematic approach for characterizing the functional “depth” of a biological neurons, suggesting that cortical pyramidal neurons and the networks they form are computationally much more powerful than previously assumed.
biorxiv neuroscience 500+-users 2019Generation of viral vectors specific to neuronal subtypes of targeted brain regions by Enhancer-Driven Gene Expression (EDGE), bioRxiv, 2019-04-14
SummaryUnderstanding brain function requires understanding neural circuits at the level of specificity at which they operate. While recent years have seen the development of a variety of remarkable molecular tools for the study of neural circuits, their utility is currently limited by the inability to deploy them in specific elements of native neural circuits, i.e. particular neuronal subtypes. One can obtain a degree of specificity with neuron-specific promoters, but native promoters are almost never sufficiently specific restricting this approach to transgenic animals. We recently showed that one can obtain transgenic mice with augmented anatomical specificity in targeted brain regions by identifying cis-regulatory elements (i.e. enhancers) uniquely active in those brain regions and combining them with a heterologous promoter, an approach we call EDGE (Enhancer-Driven Gene Expression). Here we extend this strategy to the generation of viral (rAAV) vectors, showing that when combined with the right minimal promoter they largely recapitulate the specificity seen in the corresponding transgenic lines in wildtype animals, even of another species. Because active enhancers can be identified in any tissue sample, this approach promises to enable the kind of circuit-specific manipulations in any species. This should not only greatly enhance our understanding of brain function, but may one day even provide novel therapeutic avenues to correct the imbalances in neural circuits underlying many disorders of the brain.
biorxiv neuroscience 0-100-users 2019Automated Reconstruction of a Serial-Section EM Drosophila Brain with Flood-Filling Networks and Local Realignment, bioRxiv, 2019-04-12
AbstractReconstruction of neural circuitry at single-synapse resolution is an attractive target for improving understanding of the nervous system in health and disease. Serial section transmission electron microscopy (ssTEM) is among the most prolific imaging methods employed in pursuit of such reconstructions. We demonstrate how Flood-Filling Networks (FFNs) can be used to computationally segment a forty-teravoxel whole-brain Drosophila ssTEM volume. To compensate for data irregularities and imperfect global alignment, FFNs were combined with procedures that locally re-align serial sections and dynamically adjust image content. The proposed approach produced a largely merger-free segmentation of the entire ssTEM Drosophila brain, which we make freely available. As compared to manual tracing using an efficient skeletonization strategy, the segmentation enabled circuit reconstruction and analysis workflows that were an order of magnitude faster.
biorxiv neuroscience 100-200-users 2019In the Body’s Eye The Computational Anatomy of Interoceptive Inference, bioRxiv, 2019-04-10
AbstractA growing body of evidence highlights the intricate linkage of exteroceptive perception to the rhythmic activity of the visceral body. In parallel, interoceptive inference theories of emotion and self-consciousness are on the rise in cognitive science. However, thus far no formal theory has emerged to integrate these twin domains; instead most extant work is conceptual in nature. Here, we introduce a formal model of cardiac active inference, which explains how ascending cardiac signals entrain exteroceptive sensory perception and confidence. Through simulated psychophysics, we reproduce the defensive startle reflex and commonly reported effects linking the cardiac cycle to fear perception. We further show that simulated ‘interoceptive lesions’ blunt fear expectations, induce psychosomatic hallucinations, and exacerbate metacognitive biases. Through synthetic heart-rate variability analyses, we illustrate how the balance of arousal-priors and visceral prediction errors produces idiosyncratic patterns of physiological reactivity. Our model thus offers the possibility to computationally phenotype disordered brain-body interaction.
biorxiv neuroscience 200-500-users 2019Measuring shared responses across subjects using intersubject correlation, bioRxiv, 2019-04-06
AbstractOur capacity to jointly represent information about the world underpins our social experience. By leveraging one individual’s brain activity to model another’s, we can measure shared information across brains—even in dynamic, naturalistic scenarios where an explicit response model may be unobtainable. Introducing experimental manipulations allows us to measure, for example, shared responses between speakers and listeners, or between perception and recall. In this tutorial, we develop the logic of intersubject correlation (ISC) analysis and discuss the family of neuroscientific questions that stem from this approach. We also extend this logic to spatially distributed response patterns and functional network estimation. We provide a thorough and accessible treatment of methodological considerations specific to ISC analysis, and outline best practices.
biorxiv neuroscience 100-200-users 2019Geosmin attracts Aedes aegypti mosquitoes to oviposition sites, bioRxiv, 2019-04-05
Geosmin is one of the most recognizable and common microbial smells on the planet. Some insects, like mosquitoes, require microbial-rich environments for their progeny, whereas for other insects such microbes may prove dangerous. In the vinegar fly Drosophila melanogaster, geosmin is decoded in a remarkably precise fashion and induces aversion, presumably signaling the presence of harmful microbes. We have here investigated the effect of geosmin on the behavior of the yellow fever mosquito Aedes aegypti. In contrast to flies, geosmin is not aversive in mosquitoes but stimulates egg-laying site selection. Female mosquitoes could associate geosmin with microbes, including cyanobacteria consumed by larvae, who also find geosmin – as well as geosmin producing cyanobacteria – attractive. Using in vivo multiphoton imaging from mosquitoes with pan-neural expression of the calcium reporter GCaMP6s, we show that Ae. aegypti code geosmin in a similar fashion to flies, i.e. with extreme sensitivity and with a high degree of selectivity. We further demonstrate that geosmin can be used as bait under field conditions, and finally we show that geosmin, which is both expensive and difficult to obtain, can be substituted by beetroot peel extract, providing a cheap and viable mean of mosquito control and surveillance in developing countries.
biorxiv neuroscience 0-100-users 2019