Center for Computational Biology
Center for Computational Biology
Pnevmatikakis E.A.,Center for Computational Biology |
Pnevmatikakis E.A.,Center for Theoretical Neuroscience and Grossman Center for the Statistics of Mind |
Soudry D.,Center for Theoretical Neuroscience and Grossman Center for the Statistics of Mind |
Gao Y.,Center for Theoretical Neuroscience and Grossman Center for the Statistics of Mind |
And 20 more authors.
Neuron | Year: 2016
We present a modular approach for analyzing calcium imaging recordings of large neuronal ensembles. Our goal is to simultaneously identify the locations of the neurons, demix spatially overlapping components, and denoise and deconvolve the spiking activity from the slow dynamics of the calcium indicator. Our approach relies on a constrained nonnegative matrix factorization that expresses the spatiotemporal fluorescence activity as the product of a spatial matrix that encodes the spatial footprint of each neuron in the optical field and a temporal matrix that characterizes the calcium concentration of each neuron over time. This framework is combined with a novel constrained deconvolution approach that extracts estimates of neural activity from fluorescence traces, to create a spatiotemporal processing algorithm that requires minimal parameter tuning. We demonstrate the general applicability of our method by applying it to in vitro and in vivo multi-neuronal imaging data, whole-brain light-sheet imaging data, and dendritic imaging data. Advances in calcium imaging pose significant statistical analysis challenges. Pnevmatikakis et al. present a method for identifying and spatially demixing imaged neural components and deconvolving their activity from the indicator dynamics. The method is applied to a variety of datasets. © 2016 Elsevier Inc.
Chen X.,New York University |
Yu B.,New York University |
Carriero N.,Center for Computational Biology |
Silva C.,New York University |
And 2 more authors.
Nucleic Acids Research | Year: 2017
Differential binding of transcription factors (TFs) at cis-regulatory loci drives the differentiation and function of diverse cellular lineages. Understanding the regulatory interactions that underlie cell fate decisions requires characterizing TF binding sites (TFBS) across multiple cell types and conditions. Techniques, e.g. ChIP-Seq can reveal genome-wide patterns of TF binding, but typically requires laborious and costly experiments for each TF-cell-type (TFCT) condition of interest. Chromosomal accessibility assays can connect accessible chromatin in one cell type to many TFs through sequence motif mapping. Suchmethods, however, rarely take into account that the genomic context preferred by each factor differs from TF to TF, and from cell type to cell type. To address the differences in TF behaviors, we developed Mocap, a method that integrates chromatin accessibility, motif scores, TF footprints, CpG/GC content, evolutionary conservation and other factors in an ensemble of TFCT-specific classifiers. We show that integration of genomic features, such as CpG islands improves TFBS prediction in some TFCT. Further, we describe a method for mapping new TFCT, for which no ChIP-seq data exists, onto our ensemble of classifiers and show that our cross-sample TFBS prediction method outperforms several previously described methods. © The Author(s) 2017.
Yang W.,Columbia University |
Miller J.E.K.,Columbia University |
Carrillo-Reid L.,Columbia University |
Pnevmatikakis E.,Center for Computational Biology |
And 5 more authors.
Neuron | Year: 2016
Recording the activity of large populations of neurons is an important step toward understanding the emergent function of neural circuits. Here we present a simple holographic method to simultaneously perform two-photon calcium imaging of neuronal populations across multiple areas and layers of mouse cortex in vivo. We use prior knowledge of neuronal locations, activity sparsity, and a constrained nonnegative matrix factorization algorithm to extract signals from neurons imaged simultaneously and located in different focal planes or fields of view. Our laser multiplexing approach is simple and fast, and could be used as a general method to image the activity of neural circuits in three dimensions across multiple areas in the brain. Yang et al. demonstrate a novel approach for simultaneously imaging multiple layers of the mouse cortex. They combine holographic two-photon microscopy with advanced computational source extraction to create a flexible platform for studying mesoscale neural circuits at multiple depths of the brain with cellular resolution. © 2016 Elsevier Inc.
Lee S.-J.,Center for Computational Biology |
Schlesinger P.H.,Washington University in St. Louis |
Wickline S.A.,Washington University in St. Louis |
Lanza G.M.,Washington University in St. Louis |
Baker N.A.,Pacific Northwest National Laboratory
Journal of Physical Chemistry B | Year: 2011
Melittin, an antimicrobial peptide, forms pores in biological membranes and triggers cell death. Therefore, it has potential as an anticancer therapy. However, until recently, the therapeutic application of melittin has been impractical because a suitable platform for delivery was not available. Recently, we showed that phospholipid-stabilized perfluorooctyl bromide based nanoemulsion particles (PFOB-NEPs) were resistant to destruction by melittin and enabled specific delivery of melittin to tumor cells, killing them and reducing tumor growth. Earlier, prior work also showed that melittin adsorbed onto the stabilizing phospholipid monolayer of PFOB-NEP but did not disrupt the phospholipid monolayer or produce "cracking" of the PFOB-NEPs. The present work identifies the important structural motifs for melittin binding to PFOB-NEPs through a series of atomistic molecular dynamics simulations. The conformational ensemble of melittin bound to PFOB-NEP lipid monolayer was compared to structure from a control simulation of melittin bound to a lipid bilayer to identify several differences in melittin-lipid interactions between the two systems. First, melittin was deeply buried in the hydrophobic tail region of bilayer, while its depth was attenuated in the PFOB-NEP monolayer. Second, a helical conformation was the major secondary structure in the bilayer, but the fraction of helix was reduced in the PFOB-NEP. Finally, the overall pattern for the direct interaction of melittin with surrounding lipids was similar between liposome and PFOB-NEP, but the level of interaction was slightly decreased in the PFOB-NEP. These results suggest that melittin interacts with the monolayer of PFOB-NEP in a way that is similar way to its interaction with bilayers but that deeper penetration into the hydrophobic interior is inhibited. © 2011 American Chemical Society.
News Article | October 7, 2016
A paper recently published in the Journal of Molecular Biology shows how advances in molecular biology and computer science around the world soon may lead to a three-dimensional computer model of a cell, the fundamental unit of life. According to the authors, the development could herald a new era for biological research, medical science, and human and animal health. "Cells are the foundation of life," said Ilya Vakser, professor of computational biology and molecular biosciences and director of the Center for Computational Biology at the University of Kansas, one of the paper's co-authors. "Recently, there has been tremendous progress in biomolecular modeling and advances at understanding life at the molecular level. Now, the focus is shifting to larger systems -- up to the level of the entire cell. We're trying to capture this emerging milestone development in computational structural biology, which is the tectonic shift from modeling individual biomolecular processes to modeling the entire cell." The study, titled "Challenges in structural approaches to cell modeling," surveys a range of methodologies joining the march toward a simulated whole 3-D cell, including the studies of biological networks, automated construction of 3-D cell models with experimental data, modeling of protein complexes, prediction of protein interactions, thermodynamic and kinetic effects of crowding cellular membrane modeling, and modeling of chromosomes. "A lot of techniques that are required for this are already available -- it's just a matter of putting them all together in a coherent strategy to address this problem," Vakser said. "It's hard because we're just beginning to understand the principal mechanisms of life at the molecular level -- it looks extremely complicated but doable, so we're moving very fast -- not only in our ability to understand how it works at the molecular level but to model it." While most of these techniques are being developed separately, the authors say that considered together they represent a push forward that could provide a better basic "understanding of life at the molecular level and lead to important applications to biology and medicine." "There are two major benefits," Vakser said. "One is our fundamental understanding of how a cell works. You can't claim you understand a phenomenon if you can't model it. So this gives us insight into basic fundamentals of life at the scale of an entire cell. On the practical side, it will give us an improved grasp of the underlying mechanisms of diseases and also the ability to understand mechanisms of drug action, which will be a tremendous boost to our efforts at drug design. It will help us create better drug candidates, which will potentially shorten the path to new drugs." As an example, the KU researcher said a working 3-D molecular cell model could help to replace or augment phases of time-consuming and expensive drug development protocols required today to bring drug therapies from the scientist's bench to the marketplace. Vakser said that facets of the research that could lead to a computer-simulated cell are at different levels of refinement. "We've made advances in our ability to model protein interactions," he said. "The challenge is to put it in context of the cell, which is a densely populated milieu of different proteins and other biomolecular structures. To make the transition from a dilute solution to realistic environment encountered in the cell is probably the greatest challenge we're facing right now." While modeling more complex human cells might be on the agenda soon, Vakser said that for the time being, research efforts will focus on modeling simple single-celled organisms. "We go for the simplest cell possible. There are small prokaryotic cells, which involve minimalistic set of elements that are much simpler than the bigger and more complicated cells in mammals, including humans," he said. "We're trying to cut our teeth on the smallest possible cellular organisms first, then will extrapolate into more complicated cells."
Aijo T.,Center for Computational Biology |
Aijo T.,Aalto University |
Yue X.,La Jolla Institute for Allergy and Immunology |
Rao A.,La Jolla Institute for Allergy and Immunology |
And 2 more authors.
Bioinformatics | Year: 2016
Motivation: 5-methylcytosine (5mC) is a widely studied epigenetic modification of DNA. The ten-eleven translocation (TET) dioxygenases oxidize 5mC into oxidized methylcytosines (oxi-mCs): 5-hydroxymethylcytosine (5hmC), 5-formylcytosine (5fC) and 5-carboxylcytosine (5caC). DNA methylation modifications have multiple functions. For example, 5mC is shown to be associated with diseases and oxi-mC species are reported to have a role in active DNA demethylation through 5mC oxidation and DNA repair, among others, but the detailed mechanisms are poorly understood. Bisulphite sequencing and its various derivatives can be used to gain information about all methylation modifications at single nucleotide resolution. Analysis of bisulphite based sequencing data is complicated due to the convoluted read-outs and experiment-specific variation in biochemistry. Moreover, statistical analysis is often complicated by various confounding effects. How to analyse 5mC and oxi-mC data sets with arbitrary and complex experimental designs is an open and important problem. Results: We propose the first method to quantify oxi-mC species with arbitrary covariate structures from bisulphite based sequencing data. Our probabilistic modeling framework combines a previously proposed hierarchical generative model for oxi-mC-seq data and a general linear model component to account for confounding effects. We show that our method provides accurate methylation level estimates and accurate detection of differential methylation when compared with existing methods. Analysis of novel and published data gave insights into to the demethylation of the forkhead box P3 (Foxp3) locus during the induced T regulatory cell differentiation. We also demonstrate how our covariate model accurately predicts methylation levels of the Foxp3 locus. Collectively, LuxGLM method improves the analysis of DNA methylation modifications, particularly for oxi-mC species. Availability and Implementation: An implementation of the proposed method is available under MIT license at https://github.org/tare/LuxGLM/ © 2016 The Author 2016. Published by Oxford University Press. All rights reserved.
Watkins A.M.,New York University |
Bonneau R.,Center for Computational Biology |
Arora P.S.,New York University
Journal of the American Chemical Society | Year: 2016
Protein secondary structures serve as geometrically constrained scaffolds for the display of key interacting residues at protein interfaces. Given the critical role of secondary structures in protein folding and the dependence of folding propensities on backbone dihedrals, secondary structure is expected to influence the identity of residues that are important for complex formation. Counter to this expectation, we find that a narrow set of residues dominates the binding energy in protein-protein complexes independent of backbone conformation. This finding suggests that the binding epitope may instead be substantially influenced by the side-chain conformations adopted. We analyzed side-chain conformational preferences in residues that contribute significantly to binding. This analysis suggests that preferred rotamers contribute directly to specificity in protein complex formation and provides guidelines for peptidomimetic inhibitor design. © 2016 American Chemical Society.
Salzberg S.L.,Center for Computational Biology |
Salzberg S.L.,McKusick Nathans Institute of Genetic Medicine |
Pertea M.,Center for Computational Biology |
Pertea M.,McKusick Nathans Institute of Genetic Medicine |
And 2 more authors.
Human Mutation | Year: 2014
DNA sequencing has become a powerful method to discover the genetic basis of disease. Standard, widely used protocols for analysis usually begin by comparing each individual to the human reference genome. When applied to a set of related individuals, this approach reveals millions of differences, most of which are shared among the individuals and unrelated to the disease being investigated. We have developed a novel algorithm for variant detection, one that compares DNA sequences directly to one another, without aligning them to the reference genome. When used to find de novo mutations in exome sequences from family trios, or to compare normal and diseased samples from the same individual, the new method, direct alignment for mutation discovery (DIAMUND), produces a dramatically smaller list of candidate mutations than previous methods, without losing sensitivity to detect the true cause of a genetic disease. We demonstrate our results on several example cases, including two family trios in which it correctly found the disease-causing variant while excluding thousands of harmless variants that standard methods had identified. © 2013 The Authors.
Toga A.W.,Center for Computational Biology |
Dinov I.D.,Center for Computational Biology |
Thompson P.M.,Center for Computational Biology |
Woods R.P.,Center for Computational Biology |
And 3 more authors.
Journal of the American Medical Informatics Association | Year: 2012
The Center for Computational Biology (CCB) is a multidisciplinary program where biomedical scientists, engineers, and clinicians work jointly to combine modern mathematical and computational techniques, to perform phenotypic and genotypic studies of biological structure, function, and physiology in health and disease. CCB has developed a computational framework built around the Manifold Atlas, an integrated biomedical computing environment that enables statistical inference on biological manifolds. These manifolds model biological structures, features, shapes, and flows, and support sophisticated morphometric and statistical analyses. The Manifold Atlas includes tools, workflows, and services for multimodal population-based modeling and analysis of biological manifolds. The broad spectrum of biomedical topics explored by CCB investigators include the study of normal and pathological brain development, maturation and aging, discovery of associations between neuroimaging and genetic biomarkers, and the modeling, analysis, and visualization of biological shape, form, and size. CCB supports a wide range of short-term and long-term collaborations with outside investigators, which drive the center's computational developments and focus the validation and dissemination of CCB resources to new areas and scientific domains.
PubMed | Center for Computational Biology
Type: Journal Article | Journal: Human heredity | Year: 2017
Thanks to the confluence of genomic technology and computational developments, the possibility of network inference methods that automatically learn large comprehensive models of cellular regulation is closer than ever. This perspective focuses on enumerating the elements of computational strategies that, when coupled to appropriate experimental designs, can lead to accurate large-scale models of chromatin state and transcriptional regulatory structure and dynamics. We highlight 4 research questions that require further investigation in order to make progress in network inference: (1) using overall constraints on network structure such as sparsity, (2) use of informative priors and data integration to constrain individual model parameters, (3) estimation of latent regulatory factor activity under varying cell conditions, and (4) new methods for learning and modeling regulatory factor interactions. We conclude that methods combining advances in these 4 categories of required effort with new genomic technologies will result in biophysically motivated dynamic genome-wide regulatory network models for several of the best-studied organisms and cell types.