TY - JOUR AB - A Reply to the Comment by Andrei Sergeev, M. Reizer, and V. Mitin. AU - Maksym Serbyn AU - Skvortsov, Mikhail A AU - Varlamov, Andrei A AU - Galitski, Victor M ID - 968 IS - 13 JF - Physical Review Letters TI - Serbyn et al. Reply: VL - 106 ER - TY - JOUR AB - Defining population structure and genetic diversity levels is of the utmost importance for developing efficient conservation strategies. Overfishing has caused mean annual catches of the European spiny lobster (Palinurus elephas) to decrease alarmingly along its distribution area. In this context, there is a need for comprehensive studies aiming to evaluate the genetic health of the exploited populations. The present study is based on a set of ten nuclear markers amplified in 331 individuals from ten different localities covering most of P. elephas distribution area. Samples from Atlantic and Mediterranean basins showed small but significant differences, indicating that P. elephas populations do not behave as a single panmictic unit but form two partially-overlapping groups. Despite intense overfishing, our dataset did not recover a recent bottleneck signal, and instead showed a large and stable historical effective size. This result could be accounted for by specific life-history traits (reproduction and longevity) and the limitations of molecular markers in covering recent timescales for nontemporal samples. The findings of the present study emphasize the need to integrate information on effective population sizes and life-history parameters when evaluating population connectivity levels from genetic data. AU - Palero, Ferran AU - Abello, Pere AU - Macpherson, Enrique AU - Beaumont, Mark AU - Pascual, Marta ID - 3395 IS - 2 JF - Biological Journal of the Linnean Society TI - Effect of oceanographic barriers and overfishing on the population genetic structure of the European spiny lobster Palinurus elephas VL - 104 ER - TY - GEN AB - Defining population structure and genetic diversity levels is of the utmost importance for developing efficient conservation strategies. Overfishing has caused mean annual catches of the European spiny lobster (Palinurus elephas) to decrease alarmingly along its distribution area. In this context, there is a need for comprehensive studies to evaluate the genetic health of the exploited populations. The present work is based on a set of 10 nuclear markers amplified in 331 individuals from 10 different localities covering most of P. elephas distribution area. Samples from Atlantic and Mediterranean basins showed small but significant differences, indicating that P. elephas populations do not behave as a single panmictic unit but form two partially-overlapping groups. Despite intense overfishing, our dataset did not recover a recent bottleneck signal, and showed a large and stable historical effective size instead. This result could be accounted for by specific life history traits (reproduction and longevity) and the limitations of molecular markers in covering very recent timescales for non temporal samples. Our study emphasizes the necessity of integrating information on effective population sizes and life history parameters when evaluating population connectivity levels from genetic data. AU - Palero, Ferran AU - Abello, Pere AU - Macpherson, Enrique AU - Beaumont, Mark AU - Pascual, Marta ID - 9762 TI - Data from: Effect of oceanographic barriers and overfishing on the population genetic structure of the European spiny lobster (Palinurus elephas) ER - TY - CONF AB - Segmentation is the process of partitioning digital images into meaningful regions. The analysis of biological high content images often requires segmentation as a first step. We propose ilastik as an easy-to-use tool which allows the user without expertise in image processing to perform segmentation and classification in a unified way. ilastik learns from labels provided by the user through a convenient mouse interface. Based on these labels, ilastik infers a problem specific segmentation. A random forest classifier is used in the learning step, in which each pixel's neighborhood is characterized by a set of generic (nonlinear) features. ilastik supports up to three spatial plus one spectral dimension and makes use of all dimensions in the feature calculation. ilastik provides realtime feedback that enables the user to interactively refine the segmentation result and hence further fine-tune the classifier. An uncertainty measure guides the user to ambiguous regions in the images. Real time performance is achieved by multi-threading which fully exploits the capabilities of modern multi-core machines. Once a classifier has been trained on a set of representative images, it can be exported and used to automatically process a very large number of images (e.g. using the CellProfiler pipeline). ilastik is an open source project and released under the BSD license at www.ilastik.org. AU - Sommer, Christoph M AU - Straehle, Christoph AU - Köthe, Ullrich AU - Hamprecht, Fred A. ID - 9943 KW - image segmentation KW - biomedical imaging KW - three dimensional displays KW - neurons KW - retina KW - observers KW - image color analysis SN - 1945-7928 T2 - 2011 IEEE International Symposium on Biomedical Imaging: from Nano to Micro TI - Ilastik: Interactive learning and segmentation toolkit ER - TY - CONF AB - This paper presents a method to create a model of an articulated object using the planar motion in an initialization video. The model consists of rigid parts connected by points of articulation. The rigid parts are described by the positions of salient feature-points tracked throughout the video. Following a filtering step that identifies points that belong to different objects, rigid parts are found by a grouping process in a graph pyramid. Valid articulation points are selected by verifying multiple hypotheses for each pair of parts. AU - Artner, Nicole M. AU - Ion, Adrian AU - Kropatsch, Walter G. ED - Jiang, Xiaoyi ED - Ferrer, Miquel ED - Torsello, Andrea ID - 10907 SN - 0302-9743 T2 - Graph-Based Representations in Pattern Recognition TI - Spatio-temporal extraction of articulated models in a graph pyramid VL - 6658 ER - TY - THES AB - Chemokines organize immune cell trafficking by inducing either directed (tactic) or random (kinetic) migration and by activating integrins in order to support surface adhesion (haptic). Beyond that the same chemokines can establish clearly defined functional areas in secondary lymphoid organs. Until now it is unclear how chemokines can fulfill such diverse functions. One decisive prerequisite to explain these capacities is to know how chemokines are presented in tissue. In theory chemokines could occur either soluble or immobilized, and could be distributed either homogenously or as a concentration gradient. To dissect if and how the presenting mode of chemokines influences immune cells, I tested the response of dendritic cells (DCs) to differentially displayed chemokines. DCs are antigen presenting cells that reside in the periphery and migrate into draining lymph nodes (LNs) once exposed to inflammatory stimuli to activate naïve T cells. DCs are guided to and within the LN by the chemokine receptor CCR7, which has two ligands, the chemokines CCL19 and CCL21. Both CCR7 ligands are expressed by fibroblastic reticular cells in the LN, but differ in their ability to bind to heparan sulfate residues. CCL21 has a highly charged C-terminal extension, which mediates binding to anionic surfaces, whereas CCL19 is lacking such residues and likely distributes as a soluble molecule. This study shows that surface-bound CCL21 causes random, haptokinetic DC motility, which is confined to the chemokine coated area by insideout activation of β2 integrins that mediate cell binding to the surface. CCL19 on the other hand forms concentration gradients which trigger directional, chemotactic movement, but no surface adhesion. In addition DCs can actively manipulate this system by recruiting and activating serine proteases on their surfaces, which create - by proteolytically removing the adhesive C-terminus - a solubilized variant of CCL21 that functionally resembles CCL19. By generating a CCL21 concentration gradient DCs establish a positive feedback loop to recruit further DCs from the periphery to the CCL21 coated region. In addition DCs can sense chemotactic gradients as well as immobilized haptokinetic fields at the same time and integrate these signals. The result is chemotactically biased haptokinesis - directional migration confined to a chemokine coated track or area - which could explain the dynamic but spatially tightly controlled swarming leukocyte locomotion patterns that have been observed in lymphatic organs by intravital microscopists. The finding that DCs can approach soluble cues in a non-adhesive manner while they attach to surfaces coated with immobilized cues raises the question how these cells transmit intracellular forces to the environment, especially in the non-adherent migration mode. In order to migrate, cells have to generate and transmit force to the extracellular substrate. Force transmission is the prerequisite to procure an expansion of the leading edge and a forward motion of the whole cell body. In the current conceptions actin polymerization at the leading edge is coupled to extracellular ligands via the integrin family of transmembrane receptors, which allows the transmission of intracellular force. Against the paradigm of force transmission during migration, leukocytes, like DCs, are able to migrate in threedimensional environments without using integrin transmembrane receptors (Lämmermann et al., 2008). This reflects the biological function of leukocytes, as they can invade almost all tissues, whereby their migration has to be independent from the extracellular environment. How the cells can achieve this is unclear. For this study I examined DC migration in a defined threedimensional environment and highlighted actin-dynamics with the probe Lifeact-GFP. The result was that chemotactic DCs can switch between integrin-dependent and integrin- independent locomotion and can thereby adapt to the adhesive properties of their environment. If the cells are able to couple their actin cytoskeleton to the substrate, actin polymerization is entirely converted into protrusion. Without coupling the actin cortex undergoes slippage and retrograde actin flow can be observed. But retrograde actin flow can be completely compensated by higher actin polymerization rate keeping the migration velocity and the shape of the cells unaltered. Mesenchymal cells like fibroblast cannot balance the loss of adhesive interaction, cannot protrude into open space and, therefore, strictly depend on integrinmediated force coupling. This leukocyte specific phenomenon of “adaptive force transmission” endows these cells with the unique ability to transit and invade almost every type of tissue. AU - Schumann, Kathrin ID - 3275 SN - 2663-337X TI - The role of chemotactic gradients in dendritic cell migration ER - TY - THES AU - Maître, Jean-Léon ID - 3273 SN - 2663-337X TI - Mechanics of adhesion and de‐adhesion in zebrafish germ layer progenitors ER - TY - CONF AB - We construct efficient authentication protocols and message-authentication codes (MACs) whose security can be reduced to the learning parity with noise (LPN) problem. Despite a large body of work - starting with the HB protocol of Hopper and Blum in 2001 - until now it was not even known how to construct an efficient authentication protocol from LPN which is secure against man-in-the-middle (MIM) attacks. A MAC implies such a (two-round) protocol. © 2011 International Association for Cryptologic Research AU - Kiltz, Eike AU - Pietrzak, Krzysztof Z AU - Cash, David AU - Jain, Abhishek AU - Venturi, Daniele ID - 3238 TI - Efficient authentication from hard learning problems VL - 6632 ER - TY - JOUR AB - Migrating lymphocytes acquire a polarized phenotype with a leading and a trailing edge, or uropod. Although in vitro experiments in cell lines or activated primary cell cultures have established that Rho-p160 coiled-coil kinase (ROCK)-myosin II-mediated uropod contractility is required for integrin de-adhesion on two-dimensional surfaces and nuclear propulsion through narrow pores in three-dimensional matrices, less is known about the role of these two events during the recirculation of primary, nonactivated lymphocytes. Using pharmacological antagonists of ROCK and myosin II, we report that inhibition of uropod contractility blocked integrin-independent mouse T cell migration through narrow, but not large, pores in vitro. T cell crawling on chemokine-coated endothelial cells under shear was severely impaired by ROCK inhibition, whereas transendothelial migration was only reduced through endothelial cells with high, but not low, barrier properties. Using three-dimensional thick-tissue imaging and dynamic two-photon microscopy of T cell motility in lymphoid tissue, we demonstrated a significant role for uropod contractility in intraluminal crawling and transendothelial migration through lymph node, but not bone marrow, endothelial cells. Finally, we demonstrated that ICAM-1, but not anatomical constraints or integrin-independent interactions, reduced parenchymal motility of inhibitor-treated T cells within the dense lymphoid microenvironment, thus assigning context-dependent roles for uropod contraction during lymphocyte recirculation. AU - Soriano, Silvia AU - Hons, Miroslav AU - Schumann, Kathrin AU - Kumar, Varsha AU - Dennier, Timo AU - Lyck, Ruth AU - Sixt, Michael K AU - Stein, Jens ID - 3392 IS - 5 JF - Journal of Immunology SN - 0022-1767 TI - In vivo analysis of uropod function during physiological T cell trafficking VL - 187 ER - TY - CONF AB - We study multi-label prediction for structured output sets, a problem that occurs, for example, in object detection in images, secondary structure prediction in computational biology, and graph matching with symmetries. Conventional multilabel classification techniques are typically not applicable in this situation, because they require explicit enumeration of the label set, which is infeasible in case of structured outputs. Relying on techniques originally designed for single-label structured prediction, in particular structured support vector machines, results in reduced prediction accuracy, or leads to infeasible optimization problems. In this work we derive a maximum-margin training formulation for multi-label structured prediction that remains computationally tractable while achieving high prediction accuracy. It also shares most beneficial properties with single-label maximum-margin approaches, in particular formulation as a convex optimization problem, efficient working set training, and PAC-Bayesian generalization bounds. AU - Lampert, Christoph ID - 3163 TI - Maximum margin multi-label structured prediction ER - TY - GEN AB - We study multi-label prediction for structured output spaces, a problem that occurs, for example, in object detection in images, secondary structure prediction in computational biology, and graph matching with symmetries. Conventional multi-label classification techniques are typically not applicable in this situation, because they require explicit enumeration of the label space, which is infeasible in case of structured outputs. Relying on techniques originally designed for single- label structured prediction, in particular structured support vector machines, results in reduced prediction accuracy, or leads to infeasible optimization problems. In this work we derive a maximum-margin training formulation for multi-label structured prediction that remains computationally tractable while achieving high prediction accuracy. It also shares most beneficial properties with single-label maximum-margin approaches, in particular a formulation as a convex optimization problem, efficient working set training, and PAC-Bayesian generalization bounds. AU - Lampert, Christoph ID - 3322 T2 - NIPS: Neural Information Processing Systems TI - Maximum margin multi label structured prediction ER - TY - JOUR AB - Powerful statistical models that can be learned efficiently from large amounts of data are currently revolutionizing computer vision. These models possess a rich internal structure reflecting task-specific relations and constraints. This monograph introduces the reader to the most popular classes of structured models in computer vision. Our focus is discrete undirected graphical models which we cover in detail together with a description of algorithms for both probabilistic inference and maximum a posteriori inference. We discuss separately recently successful techniques for prediction in general structured models. In the second part of this monograph we describe methods for parameter learning where we distinguish the classic maximum likelihood based methods from the more recent prediction-based parameter learning methods. We highlight developments to enhance current models and discuss kernelized models and latent variable models. To make the monograph more practical and to provide links to further study we provide examples of successful application of many methods in the computer vision literature. AU - Nowozin, Sebastian AU - Lampert, Christoph ID - 3320 IS - 3-4 JF - Foundations and Trends in Computer Graphics and Vision TI - Structured learning and prediction in computer vision VL - 6 ER - TY - CONF AB - We address the problem of metric learning for multi-view data, namely the construction of embedding projections from data in different representations into a shared feature space, such that the Euclidean distance in this space provides a meaningful within-view as well as between-view similarity. Our motivation stems from the problem of cross-media retrieval tasks, where the availability of a joint Euclidean distance function is a pre-requisite to allow fast, in particular hashing-based, nearest neighbor queries. We formulate an objective function that expresses the intuitive concept that matching samples are mapped closely together in the output space, whereas non-matching samples are pushed apart, no matter in which view they are available. The resulting optimization problem is not convex, but it can be decomposed explicitly into a convex and a concave part, thereby allowing efficient optimization using the convex-concave procedure. Experiments on an image retrieval task show that nearest-neighbor based cross-view retrieval is indeed possible, and the proposed technique improves the retrieval accuracy over baseline techniques. AU - Quadrianto, Novi AU - Lampert, Christoph ID - 3319 TI - Learning multi-view neighborhood preserving projections ER - TY - JOUR AB - The use of optical traps to measure or apply forces on the molecular level requires a precise knowledge of the trapping force field. Close to the trap center, this field is typically approximated as linear in the displacement of the trapped microsphere. However, applications demanding high forces at low laser intensities can probe the light-microsphere interaction beyond the linear regime. Here, we measured the full nonlinear force and displacement response of an optical trap in two dimensions using a dual-beam optical trap setup with back-focal-plane photodetection. We observed a substantial stiffening of the trap beyond the linear regime that depends on microsphere size, in agreement with Mie theory calculations. Surprisingly, we found that the linear detection range for forces exceeds the one for displacement by far. Our approach allows for a complete calibration of an optical trap. AU - Jahnel, Marcus AU - Behrndt, Martin AU - Jannasch, Anita AU - Schaeffer, Erik AU - Grill, Stephan ID - 3373 IS - 7 JF - Optics Letters TI - Measuring the complete force field of an optical trap VL - 36 ER - TY - JOUR AB - By exploiting an analogy between population genetics and statistical mechanics, we study the evolution of a polygenic trait under stabilizing selection, mutation and genetic drift. This requires us to track only four macroscopic variables, instead of the distribution of all the allele frequencies that influence the trait. These macroscopic variables are the expectations of: the trait mean and its square, the genetic variance, and of a measure of heterozygosity, and are derived from a generating function that is in turn derived by maximizing an entropy measure. These four macroscopics are enough to accurately describe the dynamics of the trait mean and of its genetic variance (and in principle of any other quantity). Unlike previous approaches that were based on an infinite series of moments or cumulants, which had to be truncated arbitrarily, our calculations provide a well-defined approximation procedure. We apply the framework to abrupt and gradual changes in the optimum, as well as to changes in the strength of stabilizing selection. Our approximations are surprisingly accurate, even for systems with as few as five loci. We find that when the effects of drift are included, the expected genetic variance is hardly altered by directional selection, even though it fluctuates in any particular instance. We also find hysteresis, showing that even after averaging over the microscopic variables, the macroscopic trajectories retain a memory of the underlying genetic states. AU - de Vladar, Harold AU - Barton, Nicholas H ID - 3375 IS - 58 JF - Journal of the Royal Society Interface TI - The statistical mechanics of a polygenic character under stabilizing selection mutation and drift VL - 8 ER - TY - JOUR AB - Unlike unconditionally advantageous “Fisherian” variants that tend to spread throughout a species range once introduced anywhere, “bistable” variants, such as chromosome translocations, have two alternative stable frequencies, absence and (near) fixation. Analogous to populations with Allee effects, bistable variants tend to increase locally only once they become sufficiently common, and their spread depends on their rate of increase averaged over all frequencies. Several proposed manipulations of insect populations, such as using Wolbachia or “engineered underdominance” to suppress vector-borne diseases, produce bistable rather than Fisherian dynamics. We synthesize and extend theoretical analyses concerning three features of their spatial behavior: rate of spread, conditions to initiate spread from a localized introduction, and wave stopping caused by variation in population densities or dispersal rates. Unlike Fisherian variants, bistable variants tend to spread spatially only for particular parameter combinations and initial conditions. Wave initiation requires introduction over an extended region, while subsequent spatial spread is slower than for Fisherian waves and can easily be halted by local spatial inhomogeneities. We present several new results, including robust sufficient conditions to initiate (and stop) spread, using a one-parameter cubic approximation applicable to several models. The results have both basic and applied implications. AU - Barton, Nicholas H AU - Turelli, Michael ID - 3393 IS - 3 JF - American Naturalist SN - 0003-0147 TI - Spatial waves of advance with bistable dynamics: Cytoplasmic and genetic analogues of Allee effects VL - 178 ER - TY - JOUR AB - Understanding the mechanism of protein folding requires a detailed knowledge of the structural properties of the barriers separating unfolded from native conformations. The S-peptide from ribonuclease S forms its α-helical structure only upon binding to the folded S-protein. We characterized the transition state for this binding-induced folding reaction at high resolution by determining the effect of site-specific backbone thioxylation and side-chain modifications on the kinetics and thermodynamics of the reaction, which allows us to monitor formation of backbone hydrogen bonds and side-chain interactions in the transition state. The experiments reveal that α-helical structure in the S-peptide is absent in the transition state of binding. Recognition between the unfolded S-peptide and the S-protein is mediated by loosely packed hydrophobic side-chain interactions in two well defined regions on the S-peptide. Close packing and helix formation occurs rapidly after binding. Introducing hydrophobic residues at positions outside the recognition region can drastically slow down association. AU - Bachmann, Annett AU - Wildemann, Dirk AU - Praetorius, Florian M AU - Fischer, Gunter AU - Kiefhaber, Thomas ID - 14305 IS - 10 JF - PNAS KW - Multidisciplinary SN - 0027-8424 TI - Mapping backbone and side-chain interactions in the transition state of a coupled protein folding and binding reaction VL - 108 ER - TY - JOUR AB - During assembly of the Drosophila olfactory circuit, projection neuron (PN) dendrites prepattern the developing antennal lobe before the arrival of axons from their presynaptic partners, the adult olfactory receptor neurons (ORNs). We previously found that levels of transmembrane Semaphorin-1a, which acts as a receptor, instruct PN dendrite targeting along the dorsolateral-ventromedial axis. Here we show that two secreted semaphorins, Sema-2a and Sema-2b, provide spatial cues for PN dendrite targeting. Sema-2a and Sema-2b proteins are distributed in gradients opposing the Sema-1a protein gradient, and Sema-1a binds to Sema-2a-expressing cells. In Sema-2a and Sema-2b double mutants, PN dendrites that normally target dorsolaterally in the antennal lobe mistarget ventromedially, phenocopying cell-autonomous Sema-1a removal from these PNs. Cell ablation, cell-specific knockdown, and rescue experiments indicate that secreted semaphorins from degenerating larval ORN axons direct dendrite targeting. Thus, a degenerating brain structure instructs the wiring of a developing circuit through the repulsive action of secreted semaphorins. AU - Sweeney, Lora Beatrice Jaeger AU - Chou, Ya-Hui AU - Wu, Zhuhao AU - Joo, William AU - Komiyama, Takaki AU - Potter, Christopher J. AU - Kolodkin, Alex L. AU - Garcia, K. Christopher AU - Luo, Liqun ID - 7701 IS - 5 JF - Neuron SN - 0896-6273 TI - Secreted semaphorins from degenerating larval ORN axons direct adult projection neuron dendrite targeting VL - 72 ER - TY - JOUR AB - Longitudinal axon fascicles within the Drosophila embryonic CNS provide connections between body segments and are required for coordinated neural signaling along the anterior-posterior axis. We show here that establishment of select CNS longitudinal tracts and formation of precise mechanosensory afferent innervation to the same CNS region are coordinately regulated by the secreted semaphorins Sema-2a and Sema-2b. Both Sema-2a and Sema-2b utilize the same neuronal receptor, plexin B (PlexB), but serve distinct guidance functions. Localized Sema-2b attraction promotes the initial assembly of a subset of CNS longitudinal projections and subsequent targeting of chordotonal sensory afferent axons to these same longitudinal connectives, whereas broader Sema-2a repulsion serves to prevent aberrant innervation. In the absence of Sema-2b or PlexB, chordotonal afferent connectivity within the CNS is severely disrupted, resulting in specific larval behavioral deficits. These results reveal that distinct semaphorin-mediated guidance functions converge at PlexB and are critical for functional neural circuit assembly. AU - Wu, Zhuhao AU - Sweeney, Lora Beatrice Jaeger AU - Ayoob, Joseph C. AU - Chak, Kayam AU - Andreone, Benjamin J. AU - Ohyama, Tomoko AU - Kerr, Rex AU - Luo, Liqun AU - Zlatic, Marta AU - Kolodkin, Alex L. ID - 7702 IS - 2 JF - Neuron SN - 0896-6273 TI - A combinatorial semaphorin code instructs the initial steps of sensory circuit assembly in the Drosophila CNS VL - 70 ER -