Figure - available from: Nature Methods
This content is subject to copyright. Terms and conditions apply.
![Overview of method for synaptic partner prediction, application to the FAFB dataset and use of CIRCUITMAP for circuit reconstruction and analysis in CATMAID
a, CNN predictions of postsynaptic sites (m̂\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hat{m}$$\end{document}) and direction vectors pointing to the presynaptic site (d̂\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hat{d}$$\end{document}, 3D vectors shown as RGB color) together with final detections after post-processing. Arrowheads show synaptic cleft (red), T-bar (white) and vesicles (black). b, Sample section of the FAFB dataset with predicted synaptic partners (presynaptic site, purple; postsynaptic site, turquoise). c, Using CIRCUITMAP (left), predicted synaptic partners are available in CATMAID to allow exploration of automatically reconstructed neural circuits. The example shows five automatically segmented neurons from ref. ³ (middle) together with their predicted number of synaptic connections (right; node colors match the segmentation). See also Supplementary Video 1.](publication/352719327/figure/fig3/AS:1038460253122589@1624599626632/Overview-of-method-for-synaptic-partner-prediction-application-to-the-FAFB-dataset-and.png)
Overview of method for synaptic partner prediction, application to the FAFB dataset and use of CIRCUITMAP for circuit reconstruction and analysis in CATMAID a, CNN predictions of postsynaptic sites (m̂\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hat{m}$$\end{document}) and direction vectors pointing to the presynaptic site (d̂\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\hat{d}$$\end{document}, 3D vectors shown as RGB color) together with final detections after post-processing. Arrowheads show synaptic cleft (red), T-bar (white) and vesicles (black). b, Sample section of the FAFB dataset with predicted synaptic partners (presynaptic site, purple; postsynaptic site, turquoise). c, Using CIRCUITMAP (left), predicted synaptic partners are available in CATMAID to allow exploration of automatically reconstructed neural circuits. The example shows five automatically segmented neurons from ref. ³ (middle) together with their predicted number of synaptic connections (right; node colors match the segmentation). See also Supplementary Video 1.
Source publication
We develop an automatic method for synaptic partner identification in insect brains and use it to predict synaptic partners in a whole-brain electron microscopy dataset of the fruit fly. The predictions can be used to infer a connectivity graph with high accuracy, thus allowing fast identification of neural pathways. To facilitate circuit reconstru...
Similar publications
A binary state on a graph means an assignment of binary values to its vertices. A time dependent sequence of binary states is referred to as binary dynamics. We describe a method for the classification of binary dynamics of digraphs, using particular choices of closed neighbourhoods. Our motivation and application comes from neuroscience, where a d...
Citations
... Second, in connectomics, assuming synapse strength is a function of area, it would be helpful if synapse detection computed the area of each PSD. Currently this is common for mammal synapses, but existing Drosophila synapse detectors [Huang et al., 2018, Buhmann et al., 2021 simply annotate one point within each post-synaptic density. Furthermore, these measured areas should be made available to downstream tools that use connectomic data, as well as the distance to the nearest mitochondria, and its size and darkness. ...
Mitochondria are an integral part of the metabolism of a neuron. EM images of fly brain volumes, taken for connectomics, contain mitochondria as well as the cells and synapses that have already been reported. Here, from the Drosophila hemibrain dataset, we extract, classify, and measure approximately 6 million mitochondria among roughly 21 thousand neurons of more than 5500 cell types. Each mitochondrion is classified by its appearance - dark and dense, light and sparse, or intermediate - and the location, orientation, and size (in voxels) are annotated. These mitochondria are added to our publicly available data portal, and each synapse is linked to its closest mitochondrion. Using this data, we show quantitative evidence that mitochodrial trafficing extends to the smallest dimensions in neurons. The most basic characteristics of mitochondria - volume, distance from synapses, and color - vary considerably between cell types, and between neurons with different neurotransmitters. We find that polyadic synapses with more post-synaptic densities (PSDs) have closer and larger mitochondria on the pre-synaptic side, but smaller and more distant mitochondria on the PSD side. We note that this relationship breaks down for synapses with only one PSD, suggesting a different role for such synapses.
... FlyWire provides access to a synapse table imported from Buhmann et al., 2021. We queried the table for pre-and postsynaptic sites belonging to the reconstructed sensory neurons for connectome analysis (Figure 8-figure supplements 1-3). ...
Mechanosensory neurons located across the body surface respond to tactile stimuli and elicit diverse behavioral responses, from relatively simple stimulus location-aimed movements to complex movement sequences. How mechanosensory neurons and their postsynaptic circuits influence such diverse behaviors remains unclear. We previously discovered that Drosophila perform a body location-prioritized grooming sequence when mechanosensory neurons at different locations on the head and body are simultaneously stimulated by dust (Hampel et al., 2017; Seeds et al., 2014). Here, we identify nearly all mechanosensory neurons on the Drosophila head that individually elicit aimed grooming of specific head locations, while collectively eliciting a whole head grooming sequence. Different tracing methods were used to reconstruct the projections of these neurons from different locations on the head to their distinct arborizations in the brain. This provides the first synaptic resolution somatotopic map of a head, and defines the parallel-projecting mechanosensory pathways that elicit head grooming.
... Therefore, if CL062 245 neurons drive actions in a modular manner, we would expect that different CL062 neurons will have stronger connections to different subsets of DNs. To examine whether there is modularity, we used Flywire and Natverse toolsets to obtain the connectivity of DNs that are postsynaptic of CL062 Buhmann et al., 2021;Dorkenwald et al., 2022;Heinrich et al., 2018). These connections are based on a fully reconstructed female full adult fly brain (FAFB) and curated by the flywire community (Dorkenwald 250 et al., 2022;Schlegel et al., 2023;Sven et al., 2023;Zheng et al., 2018). ...
... We utilized fafbseg to retrieve FAFB reconstructions from the production version of flywire (queried June 2023) to perform all connectomics analysis (Buhmann et al., 2021;Dorkenwald et al., 2022;Li et al., 2020;Zheng et al., 2018). We set a cutoff of 10 synapses between any two neuron pairs for all connectivity analysis and considered the number of synapses between neurons as the edge weight 745 between neurons. ...
Aggression involves both sexually monomorphic and dimorphic actions. How the brain implements these two types of actions is poorly understood. We found that a set of neurons, which we call CL062, previously shown to mediate male aggression also mediate female aggression. These neurons elicit aggression acutely and without the presence of a target. Although the same set of actions is elicited in males and females, the overall behavior is sexually dimorphic. The CL062 neurons do not express fruitless, a gene required for sexual dimorphism in flies, and expressed by most other neurons important for controlling fly aggression. Connectomic analysis suggests that these neurons have limited connections with fruitless expressing neurons that have been shown to be important for aggression, and signal to different descending neurons. Thus, CL062 is part of a monomorphic circuit for aggression that functions parallel to the known dimorphic circuits.
... Posit Software, PBC) for R (v4.2.2, R-project.org). The s-LNv neuron meshes from the hemibrain dataset (v1.2.1; Scheffer et al., 2020), and the automatic annotated presynaptic sites (Buhmann et al., 2021) were obtained from the neuprint server (neuprint.janelia.org) using the natverse libraries (v0.2.4; Bates et al., 2020) in RStudio. ...
The small ventrolateral neurons (sLNvs) are key components of the central clock in the Drosophila brain. They signal via the neuropeptide pigment‐dispersing factor (PDF) to align the molecular clockwork of different central clock neurons and to modulate downstream circuits. The dorsal terminals of the sLNvs undergo daily morphological changes that affect presynaptic sites organised by the active zone protein Bruchpilot (BRP), a homolog of mammalian ELKS proteins. However, the role of these presynaptic sites for PDF release is ill‐defined.
Here, we combined expansion microscopy with labelling of active zones by endogenously tagged BRP to examine the spatial correlation between PDF‐containing dense‐core vesicles and BRP‐labelled active zones. We found that the number of BRP‐labelled puncta in the sLNv terminals was similar while their density differed between Zeitgeber time (ZT) 2 and 14. The relative distance between BRP‐ and PDF‐labelled puncta was increased in the morning, around the reported time of PDF release. Spontaneous dense‐core vesicle release profiles of sLNvs in a publicly available ssTEM dataset (FAFB) consistently lacked spatial correlation to BRP‐organised active zones. RNAi‐mediated downregulation of brp and other active zone proteins expressed by the sLNvs did not affect PDF‐dependent locomotor rhythmicity. In contrast, down‐regulation of genes encoding proteins of the canonical vesicle release machinery, the dense‐core vesicle‐related protein CADPS, as well as PDF impaired locomotor rhythmicity.
Taken together, our study suggests that PDF release from the sLNvs is independent of BRP‐organised active zones, while BRP may be redistributed to active zones in a time‐dependent manner.
... The human electron microscopy volume was also normalized using CLAHE (contrast limited adaptive histogram equalization) 57 . We skeletonized both segmentation volumes as previously described 13,26,27,58 . ...
Maps of the nervous system that identify individual cells along with their type, subcellular components and connectivity have the potential to elucidate fundamental organizational principles of neural circuits. Nanometer-resolution imaging of brain tissue provides the necessary raw data, but inferring cellular and subcellular annotation layers is challenging. We present segmentation-guided contrastive learning of representations (SegCLR), a self-supervised machine learning technique that produces representations of cells directly from 3D imagery and segmentations. When applied to volumes of human and mouse cortex, SegCLR enables accurate classification of cellular subcompartments and achieves performance equivalent to a supervised approach while requiring 400-fold fewer labeled examples. SegCLR also enables inference of cell types from fragments as small as 10 μm, which enhances the utility of volumes in which many neurites are truncated at boundaries. Finally, SegCLR enables exploration of layer 5 pyramidal cell subtypes and automated large-scale analysis of synaptic partners in mouse visual cortex.
... It is unclear how to order experiments such that insight into whole brain dynamics are achieved efficiently. Here we show that the Flywire connectome [1][2][3][4][5][6][7] provides a feasible path to surmounting both of these barriers in the fly. ...
A long-standing goal of neuroscience is to obtain a causal model of the nervous system. This would allow neuroscientists to explain animal behavior in terms of the dynamic interactions between neurons. The recently reported whole-brain fly connectome specifies the synaptic paths by which neurons can affect each other but not whether, or how, they do affect each other in vivo. To overcome this limitation, we introduce a novel combined experimental and statistical strategy for efficiently learning a causal model of the fly brain, which we refer to as the ``effectome''. Specifically, we propose an estimator for a dynamical systems model of the fly brain that uses stochastic optogenetic perturbation data to accurately estimate causal effects and the connectome as a prior to drastically improve estimation efficiency. We then analyze the connectome to propose circuits that have the greatest total effect on the dynamics of the fly nervous system. We discover that, fortunately, the dominant circuits significantly involve only relatively small populations of neurons---thus imaging, stimulation, and neuronal identification are feasible. Intriguingly, we find that this approach also re-discovers known circuits and generates testable hypotheses about their dynamics. Overall, our analyses of the connectome provide evidence that global dynamics of the fly brain are generated by a large collection of small and often anatomically localized circuits operating, largely, independently of each other. This in turn implies that a causal model of a brain, a principal goal of systems neuroscience, can be feasibly obtained in the fly.
... Although manual annotation to analyze synapses is straightforward, it is not ideal for analysis of their density as a function of proximity to a plaque, given their large number. We thus used machine-learning-based automated synapse detection(Buhmann et al., 2021;Çiçek et al., 2016;Lin et al., 2021;Parag et al., 2019;Santurkar et al., 2017;Staffler et al., 2017;Su et al., 2023), which has been performed on large-scale connectomics datasets(Scheffer et al., 2020; Shapson-Coe et al., 2021;Turner et al., 2022) and proved to be time-efficient, unbiased and accurate. ...
Connectomics is a nascent neuroscience field to map and analyze neuronal networks. It provides a new way to investigate abnormalities in brain tissue, including in models of Alzheimer's disease (AD). This age-related disease is associated with alterations in amyloid-β (Aβ) and phosphorylated tau (pTau). These alterations correlate with AD's clinical manifestations, but causal links remain unclear. Therefore, studying these molecular alterations within the context of the local neuronal and glial milieu may provide insight into disease mechanisms. Volume electron microscopy (vEM) is an ideal tool for performing connectomics studies at the ultrastructural level, but localizing specific biomolecules within large-volume vEM data has been challenging. Here we report a volumetric correlated light and electron microscopy (vCLEM) approach using fluorescent nanobodies as immuno-probes to localize Alzheimer's disease-related molecules in a large vEM volume. Three molecules (pTau, Aβ, and a marker for activated microglia (CD11b)) were labeled without the need for detergents by three nanobody probes in a sample of the hippocampus of the 3xTg Alzheimer's disease model mouse. Confocal microscopy followed by vEM imaging of the same sample allowed for registration of the location of the molecules within the volume. This dataset revealed several ultrastructural abnormalities regarding the localizations of Aβ and pTau in novel locations. For example, two pTau-positive post-synaptic spine-like protrusions innervated by axon terminals were found projecting from the axon initial segment of a pyramidal cell. Three pyramidal neurons with intracellular Aβ or pTau were 3D reconstructed. Automatic synapse detection, which is necessary for connectomics analysis, revealed the changes in density and volume of synapse at different distances from an Aβ plaque. This vCLEM approach is useful to uncover molecular alterations within large-scale volume electron microscopy data, opening a new connectomics pathway to study Alzheimer's disease and other types of dementia.
... As peptide signals may act at a distance and may cause long-lasting neural activity state changes, studying their integration over space and time is a future challenge to further illuminate homeostatic feeding regulation. -lab, 2021;Buhmann et al., 2021;Eckstein et al., 2020;Heinrich et al., 2018 Software, algorithm R Project for Statistical Computing Dessau and Pipper, 2008 RRID:SCR_001905 ...
Consumption of food and water is tightly regulated by the nervous system to maintain internal nutrient homeostasis. Although generally considered independently, interactions between hunger and thirst drives are important to coordinate competing needs. In Drosophila , four neurons called the interoceptive subesophageal zone neurons (ISNs) respond to intrinsic hunger and thirst signals to oppositely regulate sucrose and water ingestion. Here, we investigate the neural circuit downstream of the ISNs to examine how ingestion is regulated based on internal needs. Utilizing the recently available fly brain connectome, we find that the ISNs synapse with a novel cell-type bilateral T-shaped neuron (BiT) that projects to neuroendocrine centers. In vivo neural manipulations revealed that BiT oppositely regulates sugar and water ingestion. Neuroendocrine cells downstream of ISNs include several peptide-releasing and peptide-sensing neurons, including insulin producing cells (IPCs), crustacean cardioactive peptide (CCAP) neurons, and CCHamide-2 receptor isoform RA (CCHa2R-RA) neurons. These neurons contribute differentially to ingestion of sugar and water, with IPCs and CCAP neurons oppositely regulating sugar and water ingestion, and CCHa2R-RA neurons modulating only water ingestion. Thus, the decision to consume sugar or water occurs via regulation of a broad peptidergic network that integrates internal signals of nutritional state to generate nutrient-specific ingestion.
... To assess this, the impact of technical noise on a fictive ground truth model was assessed ( Figure 4H; Methods). This model was randomly subsampled according to postsynaptic completion rate (in the mushroom body, for example, this ranges from 46% in FlyWire right to 77% in hemibrain; see Figure S4F), and synapses were randomly added and deleted according to the false-positive and -negative rates reported for the synapse detection 65 . ...
... Insect synapses are polyadic, i.e. each presynaptic site can be associated with multiple postsynaptic sites. In contrast to the Janelia hemibrain dataset, the synapse predictions used in FlyWire do not have a concept of a unitary presynaptic site associated with a T-bar 65 . Therefore, presynapse counts used in this paper do not represent the number of presynaptic sites but rather the number of outgoing connections. ...
... In the second step, we simulated the proofreading process by randomly drawing (without replacement) individual synaptic connections from the fictive ground-truth until reaching a target completion rate. We further simulate the impact of false-positive and false-negatives by randomly adding and removing synapses to/from the draw according to the precision (0.72) and recall (0.77) rates reported by Buhmann et al. 65 . In each round, we made two draws: 1. ...
The fruit fly Drosophila melanogaster combines surprisingly sophisticated behaviour with a highly tractable nervous system. A large part of the fly's success as a model organism in modern neuroscience stems from the concentration of collaboratively generated molecular genetic and digital resources. As presented in our FlyWire companion paper ¹ , this now includes the first full brain connectome of an adult animal. Here we report the systematic and hierarchical annotation of this ~130,000-neuron connectome including neuronal classes, cell types and developmental units (hemilineages). This enables any researcher to navigate this huge dataset and find systems and neurons of interest, linked to the literature through the Virtual Fly Brain database ² . Crucially, this resource includes 4,179 cell types of which 3,166 consensus cell types are robustly defined by comparison with a second dataset, the "hemibrain" connectome ³ . Comparative analysis showed that cell type counts and strong connections were largely stable, but connection weights were surprisingly variable within and across animals. Further analysis defined simple heuristics for connectome interpretation: connections stronger than 10 unitary synapses or providing >1% of the input to a target cell are highly conserved. Some cell types showed increased variability across connectomes: the most common cell type in the mushroom body, required for learning and memory, is almost twice as numerous in FlyWire than in the hemibrain. We find evidence for functional homeostasis through adjustments of the absolute amount of excitatory input while maintaining the excitation-inhibition ratio. Finally, and surprisingly, about one third of the cell types recorded in the hemibrain connectome could not be robustly identified in the FlyWire connectome, cautioning against defining cell types based on single connectomes. We propose that a cell type should be robust to inter-individual variation, and therefore defined as a group of cells that are more similar to cells in a different brain than to any other cell in the same brain. We show that this new definition can be consistently applied to whole connectome datasets. Our work defines a consensus cell type atlas for the fly brain and provides both an intellectual framework and open source toolchain for brain-scale comparative connectomics.
... Posit Software, PBC) for R (v4.2.2, R-project.org). The s-LNv neuron meshes from the hemibrain dataset (v1.2.1, (Scheffer et al., 2020)), and the automatic annotated presynaptic sites (Buhmann et al., 2021) were obtained from the neuprint server (neuprint.janelia.org) using the natverse libraries (v0.2.4, ...
The small ventrolateral neurons (sLNvs) are key components of the central clock in the Drosophila brain. They signal via the neuropeptide Pigment-dispersing factor (PDF) to align the molecular clockwork of different central clock neurons and to modulate downstream circuits. The dorsal terminals of the sLNvs undergo daily morphological changes that have been shown to affect presynaptic sites organised by the active zone protein Bruchpilot (BRP), a homolog of mammalian ELKS proteins. Although the circadian plasticity of the sLNv terminals is well established, whether and how it is related to the rhythmic release of PDF remains ill-defined.
Here, we combined expansion microscopy with labelling of active zones by endogenously tagged BRP to examine the spatial correlation between PDF-containing dense-core vesicles and BRP-labelled active zones. We found that the number of BRP-labelled punctae in the sLNv terminals remained stable while their density changed during circadian plasticity. The relative distance between BRP- and PDF-labelled punctae was increased in the morning, around the reported time of PDF release. Spontaneous dense-core vesicle release profiles of the sLNvs in a publicly available ssTEM dataset (FAFB) consistently lacked spatial correlation to BRP-organised active zones. RNAi-mediated downregulation of brp and other active zone proteins expressed by the sLNvs did not affect PDF-dependent locomotor rhythmicity. In contrast, down-regulation of genes of the canonical vesicle release machinery, the dense-core vesicle-related protein CADPS, as well as PDF impaired locomotor rhythmicity.
Taken together, our study suggests that PDF release from the sLNvs is independent of BRP-organised active zones which seem not to be circadianly destroyed and re-established.
