Search

We demonstrate gas cluster ion beam scanning electron microscopy (SEM), in which wide-area ion milling is performed on a series of thick tissue sections. This three-dimensional electron microscopy technique acquires datasets with

The neural circuits responsible for animal behavior remain largely unknown. We summarize new methods and present the circuitry of a large fraction of the brain of the fruit fly Drosophila melanogaster. Improved methods include new procedures to prepare, image, align, segment, find synapses in, and proofread such large data sets. We define cell types, refine computational compartments, and provide an exhaustive atlas of cell examples and types, many of them novel. We provide detailed circuits consisting of neurons and their chemical synapses for most of the central brain. We make the data public and simplify access, reducing the effort needed to answer circuit questions, and provide procedures linking the neurons defined by our analysis with genetic reagents. Biologically, we examine distributions of connection strengths, neural motifs on different scales, electrical consequences of compartmentalization, and evidence that maximizing packing density is an important criterion in the evolution of the fly’s brain., eLife digest Animal brains of all sizes, from the smallest to the largest, work in broadly similar ways. Studying the brain of any one animal in depth can thus reveal the general principles behind the workings of all brains. The fruit fly Drosophila is a popular choice for such research. With about 100,000 neurons – compared to some 86 billion in humans – the fly brain is small enough to study at the level of individual cells. But it nevertheless supports a range of complex behaviors, including navigation, courtship and learning. Thanks to decades of research, scientists now have a good understanding of which parts of the fruit fly brain support particular behaviors. But exactly how they do this is often unclear. This is because previous studies showing the connections between cells only covered small areas of the brain. This is like trying to understand a novel when all you can see is a few isolated paragraphs. To solve this problem, Scheffer, Xu, Januszewski, Lu, Takemura, Hayworth, Huang, Shinomiya et al. prepared the first complete map of the entire central region of the fruit fly brain. The central brain consists of approximately 25,000 neurons and around 20 million connections. To prepare the map – or connectome – the brain was cut into very thin 8nm slices and photographed with an electron microscope. A three-dimensional map of the neurons and connections in the brain was then reconstructed from these images using machine learning algorithms. Finally, Scheffer et al. used the new connectome to obtain further insights into the circuits that support specific fruit fly behaviors. The central brain connectome is freely available online for anyone to access. When used in combination with existing methods, the map will make it easier to understand how the fly brain works, and how and why it can fail to work correctly. Many of these findings will likely apply to larger brains, including our own. In the long run, studying the fly connectome may therefore lead to a better understanding of the human brain and its disorders. Performing a similar analysis on the brain of a small mammal, by scaling up the methods here, will be a likely next step along this path.

The neural circuits responsible for animal behavior remain largely unknown. We summarize new methods and present the circuitry of a large fraction of the brain of the fruit fly Drosophila melanogaster. Improved methods include new procedures to prepare, image, align, segment, find synapses in, and proofread such large data sets. We define cell types, refine computational compartments, and provide an exhaustive atlas of cell examples and types, many of them novel. We provide detailed circuits consisting of neurons and their chemical synapses for most of the central brain. We make the data public and simplify access, reducing the effort needed to answer circuit questions, and provide procedures linking the neurons defined by our analysis with genetic reagents. Biologically, we examine distributions of connection strengths, neural motifs on different scales, electrical consequences of compartmentalization, and evidence that maximizing packing density is an important criterion in the evolution of the fly’s brain.

The neural circuits responsible for behavior remain largely unknown. Previous efforts have reconstructed the complete circuits of small animals, with hundreds of neurons, and selected circuits for larger animals. Here we (the FlyEM project at Janelia and collaborators at Google) summarize new methods and present the complete circuitry of a large fraction of the brain of a much more complex animal, the fruit fly Drosophila melanogaster. Improved methods include new procedures to prepare, image, align, segment, find synapses, and proofread such large data sets; new methods that define cell types based on connectivity in addition to morphology; and new methods to simplify access to a large and evolving data set. From the resulting data we derive a better definition of computational compartments and their connections; an exhaustive atlas of cell examples and types, many of them novel; detailed circuits for most of the central brain; and exploration of the statistics and structure of different brain compartments, and the brain as a whole. We make the data public, with a web site and resources specifically designed to make it easy to explore, for all levels of expertise from the expert to the merely curious. The public availability of these data, and the simplified means to access it, dramatically reduces the effort needed to answer typical circuit questions, such as the identity of upstream and downstream neural partners, the circuitry of brain regions, and to link the neurons defined by our analysis with genetic reagents that can be used to study their functions.Note: In the next few weeks, we will release a series of papers with more involved discussions. One paper will detail the hemibrain reconstruction with more extensive analysis and interpretation made possible by this dense connectome. Another paper will explore the central complex, a brain region involved in navigation, motor control, and sleep. A final paper will present insights from the mushroom body, a center of multimodal associative learning in the fly brain.

Deriving the detailed synaptic connections of the entire nervous system has been a long term but unrealized goal of the nascent field of connectomics. For Drosophila, in particular, three sample preparation problems must be solved before the requisite imaging and analysis can even begin. The first is dissecting the brain, connectives, and ventral nerve cord (roughly comparable to the brain, neck, and spinal cord of vertebrates) as a single contiguous unit. Second is fixing and staining the resulting specimen, too large for previous techniques such as flash freezing, so as to permit the necessary automated segmentation of neuron membranes. Finally the contrast must be sufficient to support synapse detection at imaging speeds that enable the entire connectome to be collected. To address these issues, we report three major novel methods to dissect, fix, dehydrate and stain this tiny but complex nervous system in its entirety; together they enable us to uncover a Focused Ion-Beam Scanning Electron Microscopy (FIB-SEM) connectome of the entire Drosophila brain. They reliably recover fixed neurons as round profiles with darkly stained synapses, suitable for machine segmentation and automatic synapse detection, for which only minimal human intervention is required. Our advanced procedures use: a custom-made jig to microdissect both regions of the central nervous system, dorsal and ventral, with their connectives; fixation and Durcupan embedment, followed by a special hot-knife slicing protocol to reduce the brain to dimensions suited to FIB; contrast enhancement by heavy metals; together with a progressive lowering of temperature protocol for dehydration. Collectively these optimize the brain’s morphological preservation, imaging it at a usual resolution of 8nm per voxel while simultaneously speeding the formerly slow rate of FIB-SEM. With these methods we could recently obtain a FIB-SEM image stack of the Drosophila brain eight times faster than hitherto, at approximately the same rate as, but without the requirement to cut, nor imperfections in, EM serial sections.

Isotropic high-resolution imaging of large volumes provides unprecedented opportunities to advance connectomics and cell biology research. Conventional Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) offers unique benefits such as high resolution (< 10 nm in x, y, and z), robust image alignment, and minimal artifacts for superior tracing of neurites. However, its prevailing deficiencies in imaging speed and duration cap the maximum possible image volume. We have developed technologies to overcome these limitations, thereby expanding the image volume of FIB-SEM by more than four orders of magnitude from 103µm3to 3 x 107µm3while maintaining an isotropic resolution of 8 x 8 x 8 nm3voxels. These expanded volumes are now large enough to support connectomic studies, in which the superior z resolution enables automated tracing of fine neurites and reduces the time-consuming human proofreading effort. Moreover, by trading off imaging speed, the system can readily be operated at even higher resolutions achieving voxel sizes of 4 x 4 x 4 nm3, thereby generating ground truth of the smallest organelles for machine learning in connectomics and providing important insights into cell biology. Primarily limited by time, the maximum volume can be greatly extended.Here we provide a detailed description of the enhanced FIB-SEM technology, which has transformed the conventional FIB-SEM from a laboratory tool that is unreliable for more than a few days to a robust imaging platform with long term reliability: capable of years of continuous imaging without defects in the final image stack. An in-depth description of the systematic approach to optimize operating parameters based on resolution requirements and electron dose boundary conditions is also explicitly disclosed. We further explore how this technology unleashes the full potential of FIB-SEM systems, revolutionizing volume electron microscopy (EM) imaging for biology by gaining access to large sample volumes with single-digit nanoscale isotropic resolution.

We demonstrate gas cluster ion beam scanning electron microscopy (SEM), in which wide-area ion milling is performed on a series of thick tissue sections. This three-dimensional electron microscopy technique acquires datasets with 10 nm isotropic resolution of each section, and these can then be stitched together to span the sectioned volume. Incorporating gas cluster ion beam SEM into existing single-beam and multibeam SEM workflows should be straightforward, increasing reliability while improving z resolution by a factor of three or more.

Focused Ion Beam Scanning Electron Microscopy (FIB-SEM) generates 3D datasets optimally suited for segmentation of cell ultrastructure and automated connectome tracing but is limited to small fields of view and is therefore incompatible with the new generation of ultrafast multibeam SEMs. In contrast, section-based techniques are multibeam-compatible but are limited in z-resolution making automatic segmentation of cellular ultrastructure difficult. Here we demonstrate a novel 3D electron microscopy technique, Gas Cluster Ion Beam SEM (GCIB-SEM), in which top-down, wide-area ion milling is performed on a series of thick sections, acquiring < 10 nm isotropic datasets of each which are then stitched together to span the full sectioned volume. Based on our results, incorporating GCIB-SEM into existing single beam and multibeam SEM workflows should be straightforward and should dramatically increase reliability while simultaneously improving z-resolution by a factor of 3 or more.

Close appositions between the membrane of the endoplasmic reticulum (ER) and other intracellular membranes have important functions in cell physiology. These include lipid homeostasis, regulation of Ca2+ dynamics, and control of organelle biogenesis and dynamics. Although these membrane contacts have previously been observed in neurons, their distribution and abundance have not been systematically analyzed. Here, we have used focused ion beam-scanning electron microscopy to generate 3D reconstructions of intracellular organelles and their membrane appositions involving the ER (distance ≤30 nm) in different neuronal compartments. ER–plasma membrane (PM) contacts were particularly abundant in cell bodies, with large, flat ER cisternae apposed to the PM, sometimes with a notably narrow lumen (thin ER). Smaller ER–PM contacts occurred throughout dendrites, axons, and in axon terminals. ER contacts with mitochondria were abundant in all compartments, with the ER often forming a network that embraced mitochondria. Small focal contacts were also observed with tubulovesicular structures, likely to be endosomes, and with sparse multivesicular bodies and lysosomes found in our reconstructions. Our study provides an anatomical reference for interpreting information about interorganelle communication in neurons emerging from functional and biochemical studies.

Summary Using light and serial electron microscopy, we show profound refinements in motor axonal branching and synaptic connectivity before and after birth. Embryonic axons become maximally connected just before birth when they innervate ∼10-fold more muscle fibers than in maturity. In some developing muscles, axons innervate almost every muscle fiber. At birth, each neuromuscular junction is coinnervated by approximately ten highly intermingled axons (versus one in adults). Extensive die off of terminal branches occurs during the first several postnatal days, leading to much sparser arbors that still span the same territory. Despite the extensive pruning, total axoplasm per neuron increases as axons elongate, thicken, and add more synaptic release sites on their remaining targets. Motor axons therefore initially establish weak connections with nearly all available postsynaptic targets but, beginning at birth, massively redistribute synaptic resources, concentrating many more synaptic sites on many fewer muscle fibers. Analogous changes in connectivity may occur in the CNS. Video Abstract

The goal of uploading a human mind into a computer is far beyond today's technology. But exactly how far? Here I review our best cognitive and neuroscience model of the mind and show that it is well suited to provide a framework to answer this question. The model suggests that our unique "software" is mainly digital in nature and is stored redundantly in the brain's synaptic connectivity matrix (i.e., our Connectome) in a way that should allow a copy to be successfully simulated. I review the resolution necessary for extracting this Connectome and conclude that today's FIBSEM technique already meets this requirement. I then sketch out a process capable of reducing a chemically-fixed, plastic-embedded brain into a set of tapes containing 20 × 20 micron tissue pillars optimally sized for automated FIBSEM imaging, and show how these tapes could be distributed among a large number of imaging machines to accomplish the task of extracting a Connectome. The scale of such an endeavor makes it impractical, but a v...

Nonaccidental properties (NAPs) are image properties that are invariant over orientation in depth and are distinguished from metric properties (MPs) that can change continuously with variations over depth orientation. To a large extent NAPs allow facile recognition of objects at novel viewpoints. Two match-to-sample experiments with 2D or 3D appearing geons assessed sensitivity to NAP vs. MP differences. A matching geon was always identical to the sample and the distractor differed from the matching geon in either a NAP or an MP on a single generalized cone dimension. For example, if the sample was a cylinder with a slightly curved axis, the NAP distractor would have a straight axis and the MP distractor would have an axis of greater curvature than the sample. Critically, the NAP and MP differences were scaled so that the MP differences were slightly greater according to pixel energy and Gabor wavelet measures of dissimilarity. Exp. 1 used a staircase procedure to determine the threshold presentation time required to achieve 75% accuracy. Exp. 2 used a constant, brief display presentation time with reaction times and error rates as dependent measures. Both experiments revealed markedly greater sensitivity to NAP over MP differences, and this was generally true for the individual dimensions. The NAP advantage was not reflected in the similarity computations of the C2 stage of HMAX, a widely cited model of later stage cortical ventral stream processing.

Conventional heavy metal poststaining methods on thin sections lend contrast but often cause contamination. To avoid this problem, we tested several en bloc staining techniques to contrast tissue in serial sections mounted on solid substrates for examination by field emission scanning electron microscopy (FESEM). Because FESEM section imaging requires that specimens have higher contrast and greater electrical conductivity than transmission electron microscopy (TEM) samples, our technique uses osmium impregnation (OTO) to make the samples conductive while heavily staining membranes for segmentation studies. Combining this step with other classic heavy metal en bloc stains, including uranyl acetate (UA), lead aspartate, copper sulfate and lead citrate, produced clean, highly contrasted TEM and scanning electron microscopy (SEM) samples of insect, fish and mammalian nervous systems. This protocol takes 7-15 d to prepare resin-embedded tissue, cut sections and produce serial section images.

A change in the basic-level class when viewing a sequence of two objects produces a large release from adaptation in LOC compared to when the images are identical. Is this due to a change in semantics or shape? In an fMRI-adaptation experiment, subjects viewed a sequence of two objects and judged whether the stimuli were identical in shape. Different-shaped stimuli could be from the same or different basic-level classes, where the physical similarities of the pairs in the two conditions were equated by a model of simple cell similarity. BOLD responses in LOC for the two conditions were equivalent, and higher than that of the identical condition, indicating that LOC is sensitive to shape rather than to basic-level semantics.

The lateral occipital complex (LOC), a cortical region critical for human object recognition, has been shown to primarily code the shape, rather than the surface properties, of an object. But what aspects of shape? Using an fMRI-adaptation (fMRI-a) paradigm in which subjects judged whether two contour-deleted images of objects were the same or different exemplars, virtually all the adaptation in LOC [especially in LOC’s most anterior portion (pFs)] could be attributed to repetition of the parts, almost none to the repetition of local image features, such as lines or vertices, templates, or basic- or subordinate-level concepts of the object. These results support the hypothesis that the neural representation of shape in LOC is an intermediate one, encoding the parts of an object.

Focused-ion-beam scanning electron microscopy (FIB-SEM) has become an essential tool for studying neural tissue at resolutions below 10 nm x 10 nm x 10 nm, producing data sets optimized for automatic connectome tracing. We present a technical advance, ultrathick sectioning, which reliably subdivides embedded tissue samples into chunks (20 mu m thick) optimally sized and mounted for efficient, parallel FIB-SEM imaging. These chunks are imaged separately and then 'volume stitched' back together, producing a final three-dimensional data set suitable for connectome tracing.

The automated tape-collecting ultramicrotome (ATUM) makes it possible to collect large numbers of ultrathin sections quickly—the equivalent of a petabyte of high resolution images each day. However, even high throughput image acquisition strategies generate images far more slowly (at present ~1 terabyte per day). We therefore developed WaferMapper, a software package that takes a multi-resolution approach to mapping and imaging select regions within a library of ultrathin sections. This automated method selects and directs imaging of corresponding regions within each section of an ultrathin section library (UTSL) that may contain many thousands of sections. Using WaferMapper, it is possible to map thousands of tissue sections at low resolution and target multiple points of interest for high resolution imaging based on anatomical landmarks. The program can also be used to expand previously imaged regions, acquire data under different imaging conditions, or re-image after additional tissue treatments.

Patchy Insulation Myelin insulates neuronal axons such that their electrical signals travel faster and more efficiently. However, not all axons are myelinated equally. Tomassy et al. (p. 319 ; see the Perspective by Fields ) obtained detailed images from two snippets of the adult mouse brain and generated three-dimensional reconstructions of individual neurons and their myelination patterns. The images show that some axons have long, unmyelinated stretches, which might offer sites for building new connections. Thus, myelination is not an all-or-none phenomenon but rather is a characteristic of what may be a specific dialogue between the neuron and the surrounding myelin-producing cells.

SummaryThe endoplasmic reticulum (ER) often forms stacked membrane sheets, an arrangement that is likely required to accommodate a maximum of membrane-bound polysomes for secretory protein synthesis. How sheets are stacked is unknown. Here, we used improved staining and automated ultrathin sectioning electron microscopy methods to analyze stacked ER sheets in neuronal cells and secretory salivary gland cells of mice. Our results show that stacked ER sheets form a continuous membrane system in which the sheets are connected by twisted membrane surfaces with helical edges of left- or right-handedness. The three-dimensional structure of tightly stacked ER sheets resembles a parking garage, in which the different levels are connected by helicoidal ramps. A theoretical model explains the experimental observations and indicates that the structure corresponds to a minimum of elastic energy of sheet edges and surfaces. The structure allows the dense packing of ER sheets in the restricted space of a cell.

Late ventral visual areas generally consist of cells having a significant degree of translation invariance. Such a "bag of features" representation is useful for the recognition of individual objects; however, it seems unable to explain our ability to parse a scene into multiple objects and to understand their spatial relationships. We review several schemes (e.g., global features and serial attention) for how to reconcile bag-of-features representation with our ability to understand relationships; we review structural description theories that, in contrast, suggest that a neural binding mechanism assigns the features of each object in a scene to a separate "slot" to which relational information for that object is explicitly bound. Four functional magnetic resonance imaging-adaptation experiments assessed how ventral stream regions respond to rearrangements of two objects in a minimal scene that depict scene translations and relational changes. Changes of relative position (e.g., elephant above bus changing to bus above elephant) produced larger releases of adaptation in the anterior lateral occipital complex (LOC) than physically equivalent translations, providing evidence that spatial relations are explicitly encoded in the anterior LOC in agreement with structural description theories.

Recent studies have shown that machine learning can improve the accuracy of detecting object boundaries in images. In the standard approach, a boundary detector is trained by minimizing its pixel-level disagreement with human boundary tracings. This naive metric is problematic because it is overly sensitive to boundary locations. This problem is solved by metrics provided with the Berkeley Segmentation Dataset, but these can be insensitive to topo-logical differences, such as gaps in boundaries. Furthermore, the Berkeley metrics have not been useful as cost functions for supervised learning. Using concepts from digital topology, we propose a new metric called the warping error that tolerates disagreements over boundary location, penalizes topological disagreements, and can be used directly as a cost function for learning boundary detection, in a method that we call Boundary Learning by Optimization with Topological Constraints (BLOTC). We trained boundary detectors on electron microscopic images of neurons, using both BLOTC and standard training. BLOTC produced substantially better performance on a 1.2 million pixel test set, as measured by both the warping error and the Rand index evaluated on segmentations generated from the boundary labelings. We also find our approach yields significantly better segmentation performance than either gPb-OWT-UCM or multiscale normalized cut, as well as Boosted Edge Learning trained directly on our data.

We report on the development of a hardware platform for integrated tuning and closed-loop operation of MEMS gyroscopes. The platform was developed and tested for the second generation JPL/Boeing post-resonator MEMS gyroscope. The control of this device is implemented through a digital design on a field programmable gate array (FPGA). A software interface allows the user to configure, calibrate, and tune the bias voltages on the microgyro. The interface easily transitions to an embedded solution that allows for the miniaturization of the system to a single chip

Evolvable hardware is a recently proposed technology in which reconfigurable hardware under the control of an evolutionary (genetic) algorithm can automatically self-reconfigure into configurations with the desired behavior. This would not only enable the on-demand generation of new functionality when needed, but this could provide increased fault-tolerance, as the hardware would be able to cope with faults by reassigning function cells to take over the faulty ones. This paper describes ongoing work at JPL, focusing on applications to onboard image processing.

Extended abstract of a paper presented at Microscopy and Microanalysis 2011 in Nashville, Tennessee, USA, August 7–August 11, 2011.

The field of evolvable hardware or bio-inspired hardware holds the promise of automatically engineering complex electronic systems that remain adaptive during use. A growing number of experiments along these lines have been performed recently, mostly using off-the-shelf hardware or straightforward extensions of building blocks used by human engineers [1] [2]. In this paper we use the POE (Phylogeny Ontogeny Epigenesis) framework of bio-inspired hardware systems [3] and restrict evolutionary search and development considerations to pure hill-climbing search only, in order to develop some theory around evolution of electronic circuits. From this theory a new analog re-configurable hardware architecture is proposed for use in evolvable hardware. The hardware is a context switchable analog computer which can implement any general non-linear dynamic system on the level of the vector field representation. The optimization algorithm is a bio-inspired “molding” of the state-space description of the system. We call this novel hardware/ optimization algorithm platform the “modeling clay” approach to bio-inspired electronic hardware.

This paper focuses on characteristics and applications of evolvable hardware (EHW) to space systems. The motivation for looking at EHW originates in the need for more autonomous adaptive space systems. The idea of evolvable hardware becomes attractive for long missions when the hardware looses optimality, and uploading new software only partly alleviates the problem if the computing hardware becomes obsolete or the sensing hardware faces needs outside original design specifications. The paper reports the first intrinsic evolution on an analog ASIC (a custom analog neural chip), suggests evolution of dynamical systems in state-space representations, and demonstrates evolution of compression algorithms with results better than the best-known compression algorithms.

Extended abstract of a paper presented at Microscopy and Microanalysis 2006 in Chicago, Illinois, USA, July 30 – August 3, 2005