Main Menu (Mobile)- Block

Main Menu - Block

janelia7_blocks-janelia7_fake_breadcrumb | block
Lee Tzumin Lab / Publications
custom | custom

Filter

facetapi-Q2b17qCsTdECvJIqZJgYMaGsr8vANl1n | block
facetapi-W9JlIB1X0bjs93n1Alu3wHJQTTgDCBGe | block
facetapi-PV5lg7xuz68EAY8eakJzrcmwtdGEnxR0 | block
facetapi-021SKYQnqXW6ODq5W5dPAFEDBaEJubhN | block
general_search_page-panel_pane_1 | views_panes

29 Publications

Showing 1-10 of 29 results
Your Criteria:
    01/31/13 | Multi-channel acoustic recording and automated analysis of Drosophila courtship songs.
    Arthur BJ, Sunayama-Morita T, Coen P, Murthy M, Stern DL
    BMC Biology. 2013 Jan 31;11:11. doi: 10.1186/1741-7007-11-11

    Drosophila melanogaster has served as a powerful model system for genetic studies of courtship songs. To accelerate research on the genetic and neural mechanisms underlying courtship song, we have developed a sensitive recording system to simultaneously capture the acoustic signals from 32 separate pairs of courting flies as well as software for automated segmentation of songs.

    View Publication Page
    01/23/13 | Multiple interactions control synaptic layer specificity in the Drosophila visual system.
    Pecot MY, Tadros W, Nern A, Bader M, Chen Y, Zipursky SL
    Neuron. 2013 Jan 23;77(2):299-310. doi: 10.1016/j.neuron.2012.11.007

    How neurons form synapses within specific layers remains poorly understood. In the Drosophila medulla, neurons target to discrete layers in a precise fashion. Here we demonstrate that the targeting of L3 neurons to a specific layer occurs in two steps. Initially, L3 growth cones project to a common domain in the outer medulla, overlapping with the growth cones of other neurons destined for a different layer through the redundant functions of N-Cadherin (CadN) and Semaphorin-1a (Sema-1a). CadN mediates adhesion within the domain and Sema-1a mediates repulsion through Plexin A (PlexA) expressed in an adjacent region. Subsequently, L3 growth cones segregate from the domain into their target layer in part through Sema-1a/PlexA-dependent remodeling. Together, our results and recent studies argue that the early medulla is organized into common domains, comprising processes bound for different layers, and that discrete layers later emerge through successive interactions between processes within domains and developing layers.

    View Publication Page
    01/23/13 | Regularization and nonlinearities for neural language models: when are they needed?
    Pachitariu M, Sahani M
    arXiv. 2013 Jan 23:arXiv:1301.5650

    Neural language models (LMs) based on recurrent neural networks (RNN) are some of the most successful word and character-level LMs. Why do they work so well, in particular better than linear neural LMs? Possible explanations are that RNNs have an implicitly better regularization or that RNNs have a higher capacity for storing patterns due to their nonlinearities or both. Here we argue for the first explanation in the limit of little training data and the second explanation for large amounts of text data. We show state-of-the-art performance on the popular and small Penn dataset when RNN LMs are regularized with random dropout. Nonetheless, we show even better performance from a simplified, much less expressive linear RNN model without off-diagonal entries in the recurrent matrix. We call this model an impulse-response LM (IRLM). Using random dropout, column normalization and annealed learning rates, IRLMs develop neurons that keep a memory of up to 50 words in the past and achieve a perplexity of 102.5 on the Penn dataset. On two large datasets however, the same regularization methods are unsuccessful for both models and the RNN's expressivity allows it to overtake the IRLM by 10 and 20 percent perplexity, respectively. Despite the perplexity gap, IRLMs still outperform RNNs on the Microsoft Research Sentence Completion (MRSC) task. We develop a slightly modified IRLM that separates long-context units (LCUs) from short-context units and show that the LCUs alone achieve a state-of-the-art performance on the MRSC task of 60.8%. Our analysis indicates that a fruitful direction of research for neural LMs lies in developing more accessible internal representations, and suggests an optimization regime of very high momentum terms for effectively training such models.

    View Publication Page
    01/15/13 | Multidimensional traction force microscopy reveals out-of-plane rotational moments about focal adhesions.
    Legant WR, Choi CK, Miller JS, Shao L, Gao L, Betzig E, Chen CS
    Proceedings of the National Academy of Sciences of the United States of America. 2013 Jan 15;110(3):881-6. doi: 10.1073/pnas.1207997110

    Recent methods have revealed that cells on planar substrates exert both shear (in-plane) and normal (out-of-plane) tractions against the extracellular matrix (ECM). However, the location and origin of the normal tractions with respect to the adhesive and cytoskeletal elements of cells have not been elucidated. We developed a high-spatiotemporal-resolution, multidimensional (2.5D) traction force microscopy to measure and model the full 3D nature of cellular forces on planar 2D surfaces. We show that shear tractions are centered under elongated focal adhesions whereas upward and downward normal tractions are detected on distal (toward the cell edge) and proximal (toward the cell body) ends of adhesions, respectively. Together, these forces produce significant rotational moments about focal adhesions in both protruding and retracting peripheral regions. Temporal 2.5D traction force microscopy analysis of migrating and spreading cells shows that these rotational moments are highly dynamic, propagating outward with the leading edge of the cell. Finally, we developed a finite element model to examine how rotational moments could be generated about focal adhesions in a thin lamella. Our model suggests that rotational moments can be generated largely via shear lag transfer to the underlying ECM from actomyosin contractility applied at the intracellular surface of a rigid adhesion of finite thickness. Together, these data demonstrate and probe the origin of a previously unappreciated multidimensional stress profile associated with adhesions and highlight the importance of new approaches to characterize cellular forces.

    View Publication Page
    Grigorieff Lab
    01/09/13 | Location of the dsRNA-dependent polymerase, VP1, in rotavirus particles.
    Estrozi LF, Settembre EC, Goret G, McClain B, Zhang X, Chen JZ, Grigorieff N, Harrison SC
    Journal of Molecular Biology. 2013 Jan 9;425(1):124-32. doi: 10.1016/j.jmb.2012.10.011

    Double-stranded RNA (dsRNA) viruses transcribe and replicate RNA within an assembled, inner capsid particle; only plus-sense mRNA emerges into the intracellular milieu. During infectious entry of a rotavirus particle, the outer layer of its three-layer structure dissociates, delivering the inner double-layered particle (DLP) into the cytosol. DLP structures determined by X-ray crystallography and electron cryomicroscopy (cryoEM) show that the RNA coils uniformly into the particle interior, avoiding a "fivefold hub" of more structured density projecting inward from the VP2 shell of the DLP along each of the twelve 5-fold axes. Analysis of the X-ray crystallographic electron density map suggested that principal contributors to the hub are the N-terminal arms of VP2, but reexamination of the cryoEM map has shown that many features come from a molecule of VP1, randomly occupying five equivalent and partly overlapping positions. We confirm here that the electron density in the X-ray map leads to the same conclusion, and we describe the functional implications of the orientation and position of the polymerase. The exit channel for the nascent transcript directs the nascent transcript toward an opening along the 5-fold axis. The template strand enters from within the particle, and the dsRNA product of the initial replication step exits in a direction tangential to the inner surface of the VP2 shell, allowing it to coil optimally within the DLP. The polymerases of reoviruses appear to have similar positions and functional orientations.

    View Publication Page
    Svoboda Lab
    01/09/13 | Organization of cortical and thalamic input to pyramidal neurons in mouse motor cortex.
    Hooks BM, Mao T, Gutnisky DA, Yamawaki N, Svoboda K, Shepherd GM
    The Journal of Neuroscience. 2013 Jan 9;33(2):748-60. doi: 10.1523/JNEUROSCI.4338-12.2013

    Determining how long-range synaptic inputs engage pyramidal neurons in primary motor cortex (M1) is important for understanding circuit mechanisms involved in regulating movement. We used channelrhodopsin-2-assisted circuit mapping to characterize the long-range excitatory synaptic connections made by multiple cortical and thalamic areas onto pyramidal neurons in mouse vibrissal motor cortex (vM1). Each projection innervated vM1 pyramidal neurons with a unique laminar profile. Collectively, the profiles for different sources of input partially overlapped and spanned all cortical layers. Specifically, orbital cortex (OC) inputs primarily targeted neurons in L6. Secondary motor cortex (M2) inputs excited neurons mainly in L5B, including pyramidal tract neurons. In contrast, thalamocortical inputs from anterior motor-related thalamic regions, including VA/VL (ventral anterior thalamic nucleus/ventrolateral thalamic nucleus), targeted neurons in L2/3 through L5B, but avoided L6. Inputs from posterior sensory-related thalamic areas, including POm (posterior thalamic nuclear group), targeted neurons only in the upper layers (L2/3 and L5A), similar to inputs from somatosensory (barrel) cortex. Our results show that long-range excitatory inputs target vM1 pyramidal neurons in a layer-specific manner. Inputs from sensory-related cortical and thalamic areas preferentially target the upper-layer pyramidal neurons in vM1. In contrast, inputs from OC and M2, areas associated with volitional and cognitive aspects of movements, bypass local circuitry and have direct monosynaptic access to neurons projecting to brainstem and thalamus.

    View Publication Page
    01/09/13 | Unclosed HIV-1 capsids suggest a curled sheet model of assembly.
    Yu Z, Dobro MJ, Woodward CL, Levandovsky A, Danielson CM, Sandrin V, Shi J, Aiken C, Zandi R, Hope TJ, Jensen GJ
    Journal of Molecular Biology. 2013 Jan 9;425(1):112-23. doi: 10.1016/j.jmb.2012.10.006

    The RNA genome of retroviruses is encased within a protein capsid. To gather insight into the assembly and function of this capsid, we used electron cryotomography to image human immunodeficiency virus (HIV) and equine infectious anemia virus (EIAV) particles. While the majority of viral cores appeared closed, a variety of unclosed structures including rolled sheets, extra flaps, and cores with holes in the tip were also seen. Simulations of nonequilibrium growth of elastic sheets recapitulated each of these aberrations and further predicted the occasional presence of seams, for which tentative evidence was also found within the cryotomograms. To test the integrity of viral capsids in vivo, we observed that  25% of cytoplasmic HIV complexes captured by TRIM5α had holes large enough to allow internal green fluorescent protein (GFP) molecules to escape. Together, these findings suggest that HIV assembly at least sometimes involves the union in space of two edges of a curling sheet and results in a substantial number of unclosed forms.

    View Publication Page
    Tjian Lab
    01/08/13 | Dual functions of TAF7L in adipocyte differentiation.
    Zhou H, Kaplan T, Li Y, Grubisic I, Zhang Z, Wang PJ, Eisen MB, Tjian R
    eLife. 2013 Jan 8;2:e00170. doi: 10.7554/eLife.00170

    The diverse transcriptional mechanisms governing cellular differentiation and development of mammalian tissue remains poorly understood. Here we report that TAF7L, a paralogue of TFIID subunit TAF7, is enriched in adipocytes and white fat tissue (WAT) in mouse. Depletion of TAF7L reduced adipocyte-specific gene expression, compromised adipocyte differentiation, and WAT development as well. Ectopic expression of TAF7L in myoblasts reprograms these muscle precursors into adipocytes upon induction. Genome-wide mRNA-seq expression profiling and ChIP-seq binding studies confirmed that TAF7L is required for activating adipocyte-specific genes via a dual mechanism wherein it interacts with PPARγ at enhancers and TBP/Pol II at core promoters. In vitro binding studies confirmed that TAF7L forms complexes with both TBP and PPARγ. These findings suggest that TAF7L plays an integral role in adipocyte gene expression by targeting enhancers as a cofactor for PPARγ and promoters as a component of the core transcriptional machinery.DOI:http://dx.doi.org/10.7554/eLife.00170.001.

    View Publication Page
    01/08/13 | Eight pairs of descending visual neurons in the dragonfly give wing motor centers accurate population vector of prey direction.
    Gonzalez-Bellido PT, Peng H, Yang J, Georgopoulos AP, Olberg RM
    Proceedings of the National Academy of Sciences of the United States of America. 2013 Jan 8;110(2):696-701. doi: 10.1073/pnas.1210489109

    Intercepting a moving object requires prediction of its future location. This complex task has been solved by dragonflies, who intercept their prey in midair with a 95% success rate. In this study, we show that a group of 16 neurons, called target-selective descending neurons (TSDNs), code a population vector that reflects the direction of the target with high accuracy and reliability across 360°. The TSDN spatial (receptive field) and temporal (latency) properties matched the area of the retina where the prey is focused and the reaction time, respectively, during predatory flights. The directional tuning curves and morphological traits (3D tracings) for each TSDN type were consistent among animals, but spike rates were not. Our results emphasize that a successful neural circuit for target tracking and interception can be achieved with few neurons and that in dragonflies this information is relayed from the brain to the wing motor centers in population vector form.

    View Publication Page
    01/08/13 | Photon shot noise limits on optical detection of neuronal spikes and estimation of spike timing.
    Wilt BA, Fitzgerald JE, Schnitzer MJ
    Biophysical journal. 2013 Jan 08;104(1):51-62. doi: 10.1016/j.bpj.2012.07.058

    Optical approaches for tracking neural dynamics are of widespread interest, but a theoretical framework quantifying the physical limits of these techniques has been lacking. We formulate such a framework by using signal detection and estimation theory to obtain physical bounds on the detection of neural spikes and the estimation of their occurrence times as set by photon counting statistics (shot noise). These bounds are succinctly expressed via a discriminability index that depends on the kinetics of the optical indicator and the relative fluxes of signal and background photons. This approach facilitates quantitative evaluations of different indicators, detector technologies, and data analyses. Our treatment also provides optimal filtering techniques for optical detection of spikes. We compare various types of Ca(2+) indicators and show that background photons are a chief impediment to voltage sensing. Thus, voltage indicators that change color in response to membrane depolarization may offer a key advantage over those that change intensity. We also examine fluorescence resonance energy transfer indicators and identify the regimes in which the widely used ratiometric analysis of signals is substantially suboptimal. Overall, by showing how different optical factors interact to affect signal quality, our treatment offers a valuable guide to experimental design and provides measures of confidence to assess optically extracted traces of neural activity.

    View Publication Page