BBSRC MATSYB network  I2M: Immunology, Imaging and Modelling

First Workshop: 2-3 Apr 2009

School of Mathematics, University of Leeds.

[c]

Programme with abstracts

2 April (Stochastic Modelling in Biology)

  • Ernesto Estrada (University of Strathclyde)

    Communicability and Community Structure in Complex Network

    A communicability function based on walks in complex networks is introduced. This function is shown to be the thermal Green's function of the network and the concept of temperature is introduced and interpreted in this context. We obtain a spectral formula for this function for symmetric networks. The eigenvalues correspond to the energy levels of the network and the eigenvectors to vibrational modes of the nodes. Using such interpretation a clustering method is introduced, which permits to identify overlap communities in complex networks. By studying the change of temperature we analyse the existence of phase transitions in the structure of communities in complex networks.
  • Alexey Zaikin (UCL)

    Reverse engineering of proteasomal translocation rates

  • Shev MacNamara (University of Queensland/University of Oxford)

    Stochastic modeling of T Cell Homeostasis for Two Competing Clonotypes via the Master Equation

    Abstract: Stochastic models for competing clonotypes of T cells by multivariate, continuous-time, discrete state, Markov processes have been proposed in the literature by Stirk, Molina-Paris and van den Berg (2008). A stochastic modeling framework is important because of rare events associated with small populations of some critical cell types. Usually, computational methods for these problems employ a trajectory-based approach, based on Monte Carlo simulation. This is partly because the complementary, probability-density function (PDF) approaches can be expensive but in this talk we describe some efficient PDF approaches by directly solving the governing equations, known as the Master Equation. These computations are made very efficient through an approximation of the state space by the Finite State Projection and through the use of Krylov subspace methods when evolving the matrix exponential. These computational methods allow us to explore the evolution of the PDFs associated with these stochastic models, and bimodal distributions arise in some parameter regimes. Time-dependent propensities naturally arise in immunological processes due to, for example, age-dependent effects. Incorporating time-dependent propensities into the framework of the Master Equation significantly complicates the corresponding computational methods but here we describe an efficient approach via Magnus formulas. Although this contribution focuses on the example of competing clonotypes, the general principles are relevant to multivariate Markov processes and provide fundamental techniques for computational immunology.
  • Andrew Phillips (Microsoft, Cambridge)

    A Visual Programming Language for Biology

    This talk presents a visual programming language for designing and simulating computer models of biological processes. The language is based on a mathematical formalism known as the pi-calculus, and the simulation algorithm is based on standard kinetic theory of physical chemistry. The basic primitives of the language are first introduced by simple examples involving protein interactions and gene networks. The language is subsequently used to construct a model of MHC class I antigen presentation, an immune system pathway relating to the detection of foreign invaders inside a cell. The main benefit of the language is its ability to model large systems incrementally, by directly composing simpler models of subsystems.
  • Damian Clancy (University of Liverpool)

    The effect of waning immunity on long-term behaviour of infection spread

    In stochastic modelling of infectious spread, it is often assumed that infection confers permanent immunity, an SIR (susceptible-infective-removed) model. We show how results concerning long-term (endemic) behaviour may be extended to an SIRS (susceptible-infective-removed-susceptible) model, in which immunity is temporary. Our full model includes demographic processes as well as temporary immunity. We also investigate two simpler special cases: (i) An SIS model incorporating demographic processes; and (ii) an SIRS model in a closed population. Long-term behaviour is described by the quasi-stationary distribution of the process (equilibrium distribution conditional upon non-extinction of infection), which we approximate using an Ornstein-Uhlenbeck diffusion process and by a moment closure technique. We look in particular at the effect of the immune period upon infection prevalence and upon time to fade-out of infection. This is joint work with Sang Taphou Mendy.
  • Elisabeth Hultgren Hörnquist (Orebro University, Sweden) and Andreas Jansson (University of Skovde, Sweden)

    Experimental and human ulcerative colitis

    The intestinal mucosa is the largest lymphoid organ in the body, which simultaneously protects us from pathogens as well as maintains tolerance to dietary constituents and the normal intestinal flora. Dysfunctional control of local immune responses may lead to development of Inflammatory Bowel Disease(IBD) - the common name for ulcerative colitis (UC) and Crohn´s disease (CD), manifested as chronic intestinal inflammation characterized by nausea, diarrhea and severe pain. UC patients also have an increased risk of developing colorectal cancer. G proteins are a family of GTP-binding proteins that are involved in a variety of transmembrane signaling systems. Mice homozygously deficient in the G protein αi2 subunit (Gα i2-/-) develop IBD with an immunopathology similar to that observed in ulcerative colitis patients, including development of colorectal cancer. Our previous studies on the Gαi2-/- mice have demonstrated immune changes characterized by activation of proinflammatory T helper 1 cells, changes that are present, although to a lower degree, already prior to colitis development. Gαi2-/- mice respond to dietary antigens with a Th1 dominated pro-inflammatory cytokine response in the mucosa prior to colitis, whereas Gαi2+/- mice respond with increased production of the regulatory cytokine IL-10. By bone marrow and lymphocyte transfer we have demonstrated the importance of different cell population on colitis development. Our aim is to unravel the etiology of the immunopathology observed in this IBD model in the hope that we may be able to explain the mechanism/s causing disease. In addition, T cell ontogeny is studied in patients suffering from IBD. We are currently elucidating possible defects in thymocyte function and T cell ontogenesis in Gαi2-/- mice and IBD patients. Human T cell ontogeny is investigated by TRECS analysis of peripheral blood as well as mucosal lymphocytes. We are also elucidating possible defects in chemokine and chemokine receptor expression and signaling in e.g. the thymus of the Gαi2-/- mice, and its possible role in the pathogenesis of colitis.

3 April (Immunology day)

  • Anastasia Sobolewski (IFR, BBSRC)

    Bioimaging strategies to investigate microbial- immune- epithelial- cell interactions in the gut

  • Claudio Nicoletti (IFR, BBSRC)

    Dendritic cells in the gut: to sample or to exclude?

    Intestinal dendritic cells (DCs) sample bacteria, such as Salmonella by extending cellular processes into the lumen to capture bacteria and shuttle them across the epithelium; however direct evidence of bacteria-loaded DCs travelling back into the tissue is lacking. However, we have recently found that DC-mediated antigen sampling is paralleled by migration of DCs into the lumen prior to or following the internalization of Salmonella. Indeed, we found that intestinal challenge via both isolated intestinal loops and oral gavage with Salmonella ΔSPI1 induced migration of CD11c+CX3CR1+MHCII+CD11b-CD8α-DCs into the small intestine while flagellin- and SPI1-SPI2-deficient Salmonella, soluble flagellin and E. coli DH5α or flagellated K12, failed to do so. DC migration did not occur in the colon; it was not observed in MyD88 and CX3CR1 mice and intraluminal DCs internalized Salmonella but did not cross the epithelium to return into tissues. Finally, DC migration was not linked to Salmonella-induced damage of the epithelium. We interpret these data as showing that DC-mediated sampling of Salmonella is accompanied by flagellin- and MyD88-dependent migration of Salmonella-capturing DCs into the intestinal lumen. The notion that the same antigenic stimulus (e.g. infection with non-invasive Salmonella) triggers two distinct behaviours of DCs in the small bowel shows the complexity of the signalling network operating at the mucosal host-pathogen interface and it highlights the need to investigate further the role of DCs in the protection of intestinal mucosa.
  • Tim Elliott (University of Southampton)

    A multidisciplinary approach to studying immunodominance

  • Alberto Vidal-Diez (VLA, DEFRA)

    Application of Bayesian hierarchical mixture models to evaluate and optimize scrapie surveillance in Great Britain.

    Vidal-Diez1, A., Arnold, M.E1, Del Río Vilas, V.J.1,2

    1 Veterinary Laboratories Agency, Weybridge, United Kingdom
    2 Department for Environment, Food and Rural Affairs, London, United Kingdom

    The prevalence of classical scrapie, a fatal, neurological disease of small ruminants, appears to be decreasing in Great Britain. In the face of this decreasing trend there is a need to evaluate and adjust the levels of disease surveillance accordingly. Two active surveillance sources targeting individual sheep measure the occurrence of scrapie in Great Britain. As control measures are applied to holdings, there is a need to evaluate the performance of these sources to detect scrapie-affected holdings.

    Here, we present a Bayesian approach to estimate the sensitivity of the active surveillance programme of scrapie. The model is flexible to provide optimal sample sizes and designs at different levels of spatial aggregation and groups within the population of interest (e.g. holding size). More specifically, we applied a Bayesian Hierarchical Mixture Model (BHMM) to the scrapie surveillance results in 2005-2006. In general, the BHMM model consists of three components: i) a simulated population, of the sheep population clustered in holdings and flexible enough to support different levels of spatial aggregation, ii) a disease-related component that mimics the prevalence of the disease within and between holdings and iii) a surveillance module that models the operation of the surveillance sources currently in place. Parameters of the model were populated from literature, surveillance-related and demographic sources.

    Simulations were run to assess the impact of different surveillance strategies, targeting different groups within the population, on the overall sensitivity of the surveillance programme.

    Our results show that the overall sensitivity of the surveillance can be improved by sampling more sheep within holdings rather than testing more holdings and a small number of sheep within holding. Sensitivity analyses showed that the design within-holding prevalence, as imposed by the EU for sample size calculation in scrapie-affected holdings, and assumptions on the proportion of infected sheep dying on farm had the greatest impact on our results. Stratification by holding size returned a greater sensitivity than stratifying by county. The sensitivity of the AS and the FS was less than 1% and 4.85% respectively.

    We conclude that the current surveillance programme is inefficient when the objective is the detection of scrapie-affected holdings. Other potential surveillance sources, that allow a degree of control over the sampling a priori, are suggested in order to increase the level of detection of infected holdings.

    BHMM are flexible models to deal with the multiple hierarchies of animal populations and to simulate complex disease and surveillance processes. Extensions and further applications of the model in real problems are proposed.

  • Martin Turner (BI, BBSRC)

    Selection, Suicide and Survival: common aspects of lymphocyte development

  • Geoff Butcher (Babraham Institute, BBSRC)

    The GIMAP/IAN GTPases and Lymphocyte Survival

 


GDL Last revised:
Back to top of page