[Show abstract][Hide abstract] ABSTRACT: Synaptic connections between identified fast-spiking (FS), parvalbumin (PV)-positive interneurons, and excitatory spiny neurons in layer 4 (L4) of the barrel cortex were investigated using patch-clamp recordings and simultaneous biocytin fillings. Three distinct clusters of FS L4 interneurons were identified based on their axonal morphology relative to the barrel column suggesting that these neurons do not constitute a homogeneous interneuron population. One L4 FS interneuron type had an axonal domain strictly confined to a L4 barrel and was therefore named "barrel-confined inhibitory interneuron" (BIn). BIns established reliable inhibitory synaptic connections with L4 spiny neurons at a high connectivity rate of 67%, of which 69% were reciprocal. Unitary IPSPs at these connections had a mean amplitude of 0.9 ± 0.8 mV with little amplitude variation and weak short-term synaptic depression. We found on average 3.7 ± 1.3 putative inhibitory synaptic contacts that were not restricted to perisomatic areas. In conclusion, we characterized a novel type of barrel cortex interneuron in the major thalamo-recipient layer 4 forming dense synaptic networks with L4 spiny neurons. These networks constitute an efficient and powerful inhibitory feedback system, which may serve to rapidly reset the barrel microcircuitry following sensory activation.
[Show abstract][Hide abstract] ABSTRACT: Comprehensive high-resolution structural maps are central to functional exploration and understanding in biology. For the nervous system, in which high resolution and large spatial extent are both needed, such maps are scarce as they challenge data acquisition and analysis capabilities. Here we present for the mouse inner plexiform layer--the main computational neuropil region in the mammalian retina--the dense reconstruction of 950 neurons and their mutual contacts. This was achieved by applying a combination of crowd-sourced manual annotation and machine-learning-based volume segmentation to serial block-face electron microscopy data. We characterize a new type of retinal bipolar interneuron and show that we can subdivide a known type based on connectivity. Circuit motifs that emerge from our data indicate a functional mechanism for a known cellular response in a ganglion cell that detects localized motion, and predict that another ganglion cell is motion sensitive.
[Show abstract][Hide abstract] ABSTRACT: The three-dimensional (3D) structure of neural circuits is commonly studied by reconstructing individual or small groups of neurons in separate preparations. Investigation of structural organization principles or quantification of dendritic and axonal innervation thus requires integration of many reconstructed morphologies into a common reference frame. Here we present a standardized 3D model of the rat vibrissal cortex and introduce an automated registration tool that allows for precise placement of single neuron reconstructions. We (1) developed an automated image processing pipeline to reconstruct 3D anatomical landmarks, i.e., the barrels in Layer 4, the pia and white matter surfaces and the blood vessel pattern from high-resolution images, (2) quantified these landmarks in 12 different rats, (3) generated an average 3D model of the vibrissal cortex and (4) used rigid transformations and stepwise linear scaling to register 94 neuron morphologies, reconstructed from in vivo stainings, to the standardized cortex model. We find that anatomical landmarks vary substantially across the vibrissal cortex within an individual rat. In contrast, the 3D layout of the entire vibrissal cortex remains remarkably preserved across animals. This allows for precise registration of individual neuron reconstructions with approximately 30 µm accuracy. Our approach could be used to reconstruct and standardize other anatomically defined brain areas and may ultimately lead to a precise digital reference atlas of the rat brain.
[Show abstract][Hide abstract] ABSTRACT: High-resolution, comprehensive structural information is often the final arbiter between competing mechanistic models of biological processes, and can serve as inspiration for new hypotheses. In molecular biology, definitive structural data at atomic resolution are available for many macromolecules; however, information about the structure of the brain is much less complete, both in scope and resolution. Several technical developments over the past decade, such as serial block-face electron microscopy and trans-synaptic viral tracing, have made the structural biology of neural circuits conceivable: we may be able to obtain the structural information needed to reconstruct the network of cellular connections for large parts of, or even an entire, mouse brain within a decade or so. Given that the brain's algorithms are ultimately encoded by this network, knowing where all of these connections are should, at the very least, provide the data needed to distinguish between models of neural computation.
[Show abstract][Hide abstract] ABSTRACT: The connectivity architecture of neuronal circuits is essential to understand how brains work, yet our knowledge about the neuronal wiring diagrams remains limited and partial. Technical breakthroughs in labeling and imaging methods starting more than a century ago have advanced knowledge in the field. However, the volume of data associated with imaging a whole brain or a significant fraction thereof, with electron or light microscopy, has only recently become amenable to digital storage and analysis. A mouse brain imaged at light-microscopic resolution is about a terabyte of data, and 1mm(3) of the brain at EM resolution is about half a petabyte. This has given rise to a new field of research, computational analysis of large-scale neuroanatomical data sets, with goals that include reconstructions of the morphology of individual neurons as well as entire circuits. The problems encountered include large data management, segmentation and 3D reconstruction, computational geometry and workflow management allowing for hybrid approaches combining manual and algorithmic processing. Here we review this growing field of neuronal data analysis with emphasis on reconstructing neurons from EM data cubes.
Current opinion in neurobiology 02/2012; 22(1):162-9. · 7.21 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The segmentation of large volume images of neuropil acquired by serial sectioning electron microscopy is an important step toward the 3D reconstruction of neural circuits. The only cue provided by the data at hand is boundaries between otherwise indistinguishable objects. This indistinguishability, combined with the boundaries becoming very thin or faint in places, makes the large body of work on region-based segmentation methods inapplicable. On the other hand, boundary-based methods that exploit purely local evidence do not reach the extremely high accuracy required by the application domain that cannot tolerate the global topological errors arising from false local decisions. As a consequence, we propose a supervoxel merging method that arrives at its decisions in a non-local fashion, by posing and approximately solving a joint combinatorial optimization problem over all faces between supervoxels. The use of supervoxels allows the extraction of expressive geometric features. These are used by the higher-order potentials in a graphical model that assimilate knowledge about the geometry of neural surfaces by automated training on a gold standard. The scope of this improvement is demonstrated on the benchmark dataset E1088 (Helmstaedter et al., 2011) of 7.5billionvoxels from the inner plexiform layer of rabbit retina. We provide C++ source code for annotation, geometry extraction, training and inference.
Medical image analysis 12/2011; 16(4):796-805. · 3.09 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Soma location, dendrite morphology, and synaptic innervation may represent key determinants of functional responses of individual neurons, such as sensory-evoked spiking. Here, we reconstruct the 3D circuits formed by thalamocortical afferents from the lemniscal pathway and excitatory neurons of an anatomically defined cortical column in rat vibrissal cortex. We objectively classify 9 cortical cell types and estimate the number and distribution of their somata, dendrites, and thalamocortical synapses. Somata and dendrites of most cell types intermingle, while thalamocortical connectivity depends strongly upon the cell type and the 3D soma location of the postsynaptic neuron. Correlating dendrite morphology and thalamocortical connectivity to functional responses revealed that the lemniscal afferents can account for some of the cell type- and location-specific subthreshold and spiking responses after passive whisker touch (e.g., in layer 4, but not for other cell types, e.g., in layer 5). Our data provides a quantitative 3D prediction of the cell type-specific lemniscal synaptic wiring diagram and elucidates structure-function relationships of this physiologically relevant pathway at single-cell resolution.
[Show abstract][Hide abstract] ABSTRACT: How does the brain compute? Answering this question necessitates neuronal connectomes, annotated graphs of all synaptic connections within defined brain areas. Further, understanding the energetics of the brain's computations requires vascular graphs. The assembly of a connectome requires sensitive hardware tools to measure neuronal and neurovascular features in all three dimensions, as well as software and machine learning for data analysis and visualization. We present the state of the art on the reconstruction of circuits and vasculature that link brain anatomy and function. Analysis at the scale of tens of nanometers yields connections between identified neurons, while analysis at the micrometer scale yields probabilistic rules of connection between neurons and exact vascular connectivity.
Journal of Neuroscience 11/2011; 31(45):16125-38. · 6.91 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Although physiological data on microcircuits involving a few inhibitory neurons in the mammalian cerebral cortex are available, data on the quantitative relation between inhibition and excitation in cortical circuits involving thousands of neurons are largely missing. Because the distribution of neurons is very inhomogeneous in the cerebral cortex, it is critical to map all neurons in a given volume rather than to rely on sparse sampling methods. Here, we report the comprehensive mapping of interneurons (INs) in cortical columns of rat somatosensory cortex, immunolabeled for neuron-specific nuclear protein and glutamate decarboxylase. We found that a column contains ~2,200 INs (11.5% of ~19,000 neurons), almost a factor of 2 less than previously estimated. The density of GABAergic neurons was inhomogeneous between layers, with peaks in the upper third of L2/3 and in L5A. IN density therefore defines a distinct layer 2 in the sensory neocortex. In addition, immunohistochemical markers of IN subtypes were layer-specific. The "hot zones" of inhibition in L2 and L5A match the reported low stimulus-evoked spiking rates of excitatory neurons in these layers, suggesting that these inhibitory hot zones substantially suppress activity in the neocortex.
Proceedings of the National Academy of Sciences 09/2011; 108(40):16807-12. · 9.81 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The proper connectivity between neurons is essential for the implementation of the algorithms used in neural computations, such as the detection of directed motion by the retina. The analysis of neuronal connectivity is possible with electron microscopy, but technological limitations have impeded the acquisition of high-resolution data on a large enough scale. Here we show, using serial block-face electron microscopy and two-photon calcium imaging, that the dendrites of mouse starburst amacrine cells make highly specific synapses with direction-selective ganglion cells depending on the ganglion cell's preferred direction. Our findings indicate that a structural (wiring) asymmetry contributes to the computation of direction selectivity. The nature of this asymmetry supports some models of direction selectivity and rules out others. It also puts constraints on the developmental mechanisms behind the formation of synaptic connections. Our study demonstrates how otherwise intractable neurobiological questions can be addressed by combining functional imaging with the analysis of neuronal connectivity using large-scale electron microscopy.
[Show abstract][Hide abstract] ABSTRACT: Neuroanatomic analysis depends on the reconstruction of complete cell shapes. High-throughput reconstruction of neural circuits, or connectomics, using volume electron microscopy requires dense staining of all cells, which leads even experts to make annotation errors. Currently, reconstruction speed rather than acquisition speed limits the determination of neural wiring diagrams. We developed a method for fast and reliable reconstruction of densely labeled data sets. Our approach, based on manually skeletonizing each neurite redundantly (multiple times) with a visualization-annotation software tool called KNOSSOS, is ∼50-fold faster than volume labeling. Errors are detected and eliminated by a redundant-skeleton consensus procedure (RESCOP), which uses a statistical model of how true neurite connectivity is transformed into annotation decisions. RESCOP also estimates the reliability of consensus skeletons. Focused reannotation of difficult locations promises a rather steep increase of reliability as a function of the average skeleton redundancy and thus the nearly error-free analysis of large neuroanatomical datasets.
[Show abstract][Hide abstract] ABSTRACT: This is the second article in a series of three studies that investigate the anatomical determinants of thalamocortical (TC) input to excitatory neurons in a cortical column of rat primary somatosensory cortex (S1). Here, we report the number and distribution of NeuN-positive neurons within the C2, D2, and D3 TC projection columns in P27 rat somatosensory barrel cortex based on an exhaustive identification of 89,834 somata in a 1.15 mm(3) volume of cortex. A single column contained 19,109 ± 444 neurons (17,560 ± 399 when normalized to a standard-size projection column). Neuron density differences along the vertical column axis delineated "cytoarchitectonic" layers. The resulting neuron numbers per layer in the average column were 63 ± 10 (L1), 2039 ± 524 (L2), 3735 ± 905 (L3), 4447 ± 439 (L4), 1737 ± 251 (L5A), 2235 ± 99 (L5B), 3786 ± 168 (L6A), and 1066 ± 170 (L6B). These data were then used to derive the layer-specific action potential (AP) output of a projection column. The estimates confirmed previous reports suggesting that the ensembles of spiny L4 and thick-tufted pyramidal neurons emit the major fraction of APs of a column. The number of APs evoked in a column by a sensory stimulus (principal whisker deflection) was estimated as 4441 within 100 ms post-stimulus.
[Show abstract][Hide abstract] ABSTRACT: This is the concluding article in a series of 3 studies that investigate the anatomical determinants of thalamocortical (TC) input to excitatory neurons in a cortical column of rat primary somatosensory cortex (S1). We used viral synaptophysin-enhanced green fluorescent protein expression in thalamic neurons and reconstructions of biocytin-labeled cortical neurons in TC slices to quantify the number and distribution of boutons from the ventral posterior medial (VPM) and posteromedial (POm) nuclei potentially innervating dendritic arbors of excitatory neurons located in layers (L)2-6 of a cortical column in rat somatosensory cortex. We found that 1) all types of excitatory neurons potentially receive substantial TC input (90-580 boutons per neuron); 2) pyramidal neurons in L3-L6 receive dual TC input from both VPM and POm that is potentially of equal magnitude for thick-tufted L5 pyramidal neurons (ca. 300 boutons each from VPM and POm); 3) L3, L4, and L5 pyramidal neurons have multiple (2-4) subcellular TC innervation domains that match the dendritic compartments of pyramidal cells; and 4) a subtype of thick-tufted L5 pyramidal neurons has an additional VPM innervation domain in L4. The multiple subcellular TC innervation domains of L5 pyramidal neurons may partly explain their specific action potential patterns observed in vivo. We conclude that the substantial potential TC innervation of all excitatory neuron types in a cortical column constitutes an anatomical basis for the initial near-simultaneous representation of a sensory stimulus in different neuron types.
[Show abstract][Hide abstract] ABSTRACT: 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 topological 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.
The Twenty-Third IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2010, San Francisco, CA, USA, 13-18 June 2010; 01/2010
[Show abstract][Hide abstract] ABSTRACT: The axonal projections of nerve cells have been used to infer synaptic connectivity ever since the drawings of Ramon y Cajal
more than a hundred years ago. Here we review the assumptions behind these studies and report how axonal projections of thalamic
and cortical neurons can be used to anatomically define cortical columns as innervation volumes in rat barrel cortex. We then
apply this analysis to cortical interneurons and illustrate that it is the axonal projections of interneurons which best permit
their functional classification with reference to cortical columns. We conclude that the axons of cortical nerve cells should
serve as their primary classifiers, because they best indicate their function in the neocortical neuronal network.
Keywordsaxons-interneurons-cortical columns-barrel cortex-innervation-synaptic connection-innervation-thalamocortical-cluster analysis
[Show abstract][Hide abstract] ABSTRACT: Images can be segmented by first using a classifier to predict an affinity graph that reflects the degree to which image pixels must be grouped together and then partitioning the graph to yield a segmentation. Machine learning has been applied to the affinity classifier to produce affinity graphs that are good in the sense of minimizing edge misclassification rates. However, this error measure is only indirectly related to the quality of segmentations produced by ultimately partitioning the affinity graph. We present the first machine learning algorithm for training a classifier to produce affinity graphs that are good in the sense of producing segmentations that directly minimize the Rand index, a well known segmentation performance measure. The Rand index measures segmentation performance by quantifying the classification of the connectivity of image pixel pairs after segmentation. By using the simple graph partitioning algorithm of finding the connected components of the thresholded affinity graph, we are able to train an affinity classifier to directly minimize the Rand index of segmentations resulting from the graph partitioning. Our learning algorithm corresponds to the learning of maximin affinities between image pixel pairs, which are predictive of the pixel-pair connectivity.
[Show abstract][Hide abstract] ABSTRACT: Many image segmentation algorithms first generate an affinity graph and then partition it. We present a machine learning approach to computing an affinity graph using a convolutional network (CN) trained using ground truth provided by human experts. The CN affinity graph can be paired with any standard partitioning algorithm and improves segmentation accuracy significantly compared to standard hand-designed affinity functions. We apply our algorithm to the challenging 3D segmentation problem of reconstructing neuronal processes from volumetric electron microscopy (EM) and show that we are able to learn a good affinity graph directly from the raw EM images. Further, we show that our affinity graph improves the segmentation accuracy of both simple and sophisticated graph partitioning algorithms. In contrast to previous work, we do not rely on prior knowledge in the form of hand-designed image features or image preprocessing. Thus, we expect our algorithm to generalize effectively to arbitrary image types.
[Show abstract][Hide abstract] ABSTRACT: Recent technological developments have renewed the interest in large-scale neural circuit reconstruction. To resolve the structure of entire circuits, thousands of neurons must be reconstructed and their synapses identified. Reconstruction techniques at the light microscopic level are capable of following sparsely labeled neurites over long distances, but fail with densely labeled neuropil. Electron microscopy provides the resolution required to resolve densely stained neuropil, but is challenged when data for volumes large enough to contain complete circuits need to be collected. Both photon-based and electron-based imaging methods will ultimately need highly automated data analysis, because the manual tracing of most networks of interest would require hundreds to tens of thousands of years in human labor.
Current opinion in neurobiology 05/2009; 18(6):633-41. · 7.21 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: In the neocortex, inhibition by gamma-aminobutyric acidergic (GABAergic) interneurons is essential for shaping cortical maps, which represent sensory signals. For a detailed understanding of the stream of excitation evoked, for example, by a sensory stimulus, interneurons must be identified with reference to their impact on excitatory neurons located in different laminae of the same (home) and surround columns. We analyzed the axonal projection of layer 2/3 (L2/3) interneurons with reference to geometric landmarks of cortical columns by staining neurons in acute slices of rat barrel cortex (P20-P29) and a subsequent cluster analysis using morphological parameters that described the spatial distribution of axons. The cluster analysis defined 4 main axonal projection "types" referred to as 1) "local inhibitors" (including "chandelier neurons"), 2) "lateral inhibitors," 3) "translaminar L2/3-to-L4/5 inhibitors," and 4) "translaminar L2/3-to-L1 inhibitors." The putative innervation domains established by axonal projections of the 4 types of interneurons and the dendritic domains of their target excitatory neurons were 1) L2/3 of the home column, 2) L2/3 of both the home and neighboring columns, 3) L4 and L5A of the home column, and 4) L1 and L2/3 of the home column. The quantitative analysis of the axonal projection patterns of an unselected sample of 51 interneurons located in L2/3 thus defined anatomical correlates for local, lateral, and translaminar inhibition within and between cortical columns