Joint Applied Mathematics and Medical School Seminar
Bioinformatics Laboratory of Wadsworth Center and Department of Computer Science at Rensselaer Polytechnic Institute | |
A Transcription Regulation Example | |
Abstract: In the traditional hypothesis-driven biomedical research paradigm, the design of experiments to address specific prior hypotheses takes center stage. But with the genome era has come a large mass of fundamental biological data gathered without prior hypothesis, spawning a new "data-driven" paradigm. The coupling of this paradigm with high throughput experimental technologies promise to greatly accelerate biomedical research. Studies of transcription regulation in E.coli provide an example of the potential of this approach. Elucidating the transcription regulatory networks of species is a grand challenge of the post-genomic era. Toward that end we recently applied Bayesian algorithms with the goal of locating cis-regulatory sites via cross species comparison in proteobacteria (McCue et.al, NAR, 2001). Application of these technologies to a study set of 184 E. coli genes revealed that 75% of our predictions overlap documented transcription regulatory sites by \leq 10 bp. That the remaining predictions included bona fide TF binding sites was demonstrated by affinity purification of a putative transcription factor (YijC) bound to predicted but undocumented sites upstream of the fabA, fabB, and yqfA genes. Through application to the complete set of intergenic regions in E. coli, regulatory sites for 2097 genes were predicted and are available at http://www.wadsworth.org/resnres/bioinfo/. These sites represent a set of testable hypotheses. The challenge now is to scale up validation from three oligomers to thousands. The emergence of syntenic sequence from multiple vertebrate species offers similar opportunities for these species. I'll also report on identification of cis-regulatory elements based on recent extensions for the human/mouse sequence comparison technology described by Wasserman et.al., (Nature Genetics, 2000).
Brown University Center for Statistical Sciences Seminar
Abstract: The objective of this talk is to highlight the role that statistical models can play in our understanding of the recent anthrax outbreaks. Statistical ideas and reasoning can be used to estimate the incubation period of anthrax, estimate dates of exposure and estimate numbers of cases that may have been prevented by the public health response. Statistical issues include issues in left truncation, likelihood methods and sensitivity analyses. This talk represents ongoing work. Specific results relevant to the anthrax outbreaks in the United States last fall will be discussed.
Brown University Joint Mechanics/Materials Seminar
Mechanical, Chemical, & Bio-Engineering, Structural Biology Program, The Wistar Institute, Laboratory for Research on the Structure of Matter, University of Pennsylvania | |
Abstract: Although living systems have had eons to evolve, cell membranes across diverse phyla share many common structural principles and motifs. These include interfacially-driven self-assemblies and protein networks, plus mechanisms of cell adhesion/repulsion. Basic mechanical studies, from single molecule extension using Atomic Force Microscopy [1] to cellular scale using micromanipulation-coupled methods [2-5], will be presented in overview for select bilmembrane systems and their critical componentry. Complementary simulations at both the nano- and micro- scale provide a quantitative basis for further hypothesis testing. In the spirit of understanding nature's systems even better and with an eye toward engineering novel delivery systems, synthetic polymer mimics of biomembranes will also be outlined [6,7]. (www.seas.upenn.edu/~discher)
Brown University Graduate School Dissertation Defense
Brown Analysis Seminar
Brown University Graduate School Dissertation Defense
Brown University Graduate School Dissertation Defense
Brown Special Analysis Seminar
Abstract: To approximate all roots (zeros) of a univariate polynomial, we develop two effective algorithms and combine them in a single recursive process. One algorithm computes a basic well isolated zero-free annulus on the complex plane, whereas another algorithm numerically splits the input polynomial of the $n$-th degree into two factors balanced in the degrees and with the zero sets separated by the basic annulus. Recursive combination of the two algorithms leads to computation of the complete numerical factorization of a polynomial into the product of linear factors and further to the approximation of the roots. The new root-finder incorporates the earlier techniques of Sch\"onhage, Neff/Reif, and Kirrinnis and our old and new techniques and yields nearly optimal (up to polylogarithmic factors) arithmetic and Boolean cost estimates for the computational complexity of both complete factorization and root-finding. The improvement over our previous record Boolean complexity estimates is by roughly the factor of $n$ for complete factorization and also for the approximation of well-conditioned (well isolated) roots, whereas the same algorithm is also optimal (under both arithmetic and Boolean models of computing) for the worst case input polynomial, whose roots can be ill-conditioned, forming clusters. (The worst case complexity bounds for root-finding are supported by our previous algorithms as well.) All algorithms allow processor efficient acceleration to achieve solution in polylogarithmic parallel time.
Applied Mathematics Colloquium
Scientific Computing Seminar
Virginia Tech, Virginia | |
Abstract: In the past few years, there has been considerable interest in uncertainty analysis and management among the computational fluid dynamics (CFD) community. One of the primary drivers for this is NASA's risk-based design methodologies which require a multi-disciplinary, probabilistic framework. Historically, the structures and dynamics and controls disciplines have a long history of applying uncertainty methods, whereas the CFD community is a relative newcomer due, in part, to the large cost of deterministic CFD simulations and, in part, to the relative infancy of the discipline. This seminar provides an overview of some selected uncertainty methods with applications to fundamental problems in fluid mechanics. Probabilistic (Monte-Carlo, Moment Methods, Polynomial Chaos) and non-probabilistic methods (Interval Analysis, Propagation of error using sensitivity derivatives) are described. Results are presented for a model convection equation with a source term, a non-linear convection-diffusion equation, supersonic flow over wedges, expansion corners, and an airfoil; and a two-dimensional boundary layer flow.
Brown University Graduate School Disseration Defense
PDE Seminar
<--- 2002 Index