Penn Symposium on Mathematical & Computational Biology May 22-23 , 2017
K. Burrage,B. Rodríguez,A. Chakraborty
Abstract:Oscillations represent a phenomenon of temporal self-organization that arises from feedback processes in cellular regulatory networks. After providing an overview of oscillations in cellular systems and of their underlying mechanisms, I will focus on the cell cycle, which provides a major example of cellular rhythm. The mammalian cell cycle is driven by a network of cyclin-dependent kinases (CDKs). Computational models show how the regulatory design of the CDK network results in its temporal self-organization in the form of sustained oscillations that bring about the orderly progression along cell cycle phases. Multiple factors, intrinsic or extrinsic to the CDK network, control the balance between cell cycle arrest and cell proliferation, which is often deregulated in cancer. Among these factors are oncogenes, tumor suppressors, growth factors, as well as the extracellular matrix, and contact inhibition, which increases with cell density. Supra-threshold changes in the level of any of these factors can trigger a switch in the dynamical behavior of the CDK network between a stable steady state, associated with cell cycle arrest, and sustained CDK oscillations, associated with cell proliferation. That different oscillatory processes may interact at the cellular level is illustrated by the coupling of the cell cycle to the circadian clock, which results in the synchronization of these two major cellular rhythms. Shishi Luo – University of California, Berkeley The evolution of the IGHV and TRBV gene families Abstract: Despite the crucial role played by the immunoglobulin heavy variable (IGHV) and T cell beta variable (TRBV) loci in adaptive immune function, their genetic variation in the human population remains poorly characterized. Generated through a process of gene duplication/deletion and diversification, these loci can vary between individuals in gene copy number and contain genes that are highly similar. Such characteristics make the analysis of these loci technically challenging. Here, using a customized bioinformatic analysis, we quantify the copy number and single nucleotide variation in these two regions in a globally diverse sample of hundreds of individuals sequenced by the Simons Genome Diversity Project. We find that despite the shared molecular and functional characteristics between the IGHV and TRBV gene families, they exhibit starkly different patterns of variation. In particular, we find that the IGHV locus has a greater propensity for gene duplication and deletion, as evidenced by a high level of copy number variation, while the TRBV seems to have a greater propensity for nucleotide mutations, as evidenced by a moderately high level of allelic diversity. This pattern appears to be consistent across other species and suggests that the IGHV and TRBV loci have followed different evolutionary paths in their 500 million years or so of existence. Erick Matsen – Fred Hutchinson Cancer Research Center Learning about antibody affinity maturation from sequence data using probabilistic models Abstract: Our survival in the face of pathogen exposure depends on a carefully orchestrated evolutionary process acting on our B lymphocytes. This process, called “affinity maturation,” is driven by specialized machinery to mutate and enforce selection pressure on the B cell receptor (BCR) sequences that encode antibody specificity. It has recently become possible to sequence BCRs in high throughput. Although these sequences implicitly contain a wealth of information about both antigen exposure and the process by which we learn to resist pathogens, statistical methods and computer algorithms are needed to extract this information. In this talk, I will describe two recent projects to develop model-based inferential tools for analyzing BCR sequences: first, a survival model framework to learn the nucleotide context sensitivity of the mutation process in affinity maturation, and second, a branching process model that allows us to improve phylogenetic estimation by using sequence abundances. This work is joint with David Shaw and Will DeWitt (Fred Hutch), and Jean Feng, Vladimir Minin, and Noah Simon (University of Washington) Thierry Mora – École Normale Supérieure Optimal immune systems Abstract: Immune systems protect us against a large diversity of pathogens, while being tolerant to our own molecules. What are the general design principles of immune systems that allow them to perform this complex task efficiently and reliably, with the physical constraints under which they operate? I will explore this question by considering optimal immune designs at three different scales of organisation. At the molecular scale, how do receptors and signaling network distinguish between foreign and self ligands? At the individual scale, how should the repertoire of specificities be organised to best detect infections? And at the population level, what are the optimal modes of immunity (adaptive, CRISPR, etc.) as a function of the pathogen statistics? Eric Siggia – Rockefeller University Computation to study insect evolution and temperature compensation in circadian clocks Abstract: Computational evolution by mutation-selection is used to derive gene regulatory networks that perform a variety of functions. Recent examples formulate an analogy with the transition state in a chemical reaction to derive the anterior-posterior patterning in the last common ancestor between fly and mosquito. The circadian clock is built from temperature dependent components, yet has a temperature independent period. A possible architecture was derived computationally and was recently shown to have experimental support. Christine Vogel – New York University The ups and downs of protein expression regulation Abstract: Gene expression is regulated by four major processes: transcription, translation, and RNA and protein degradation. These processes are adjusted, in different ways, when the cells respond to a stimulus. Many pathways are known, but their precise interaction over time is not well understood. In our lab, we use multiple time series datasets — on protein and mRNA expression changes and changes in the binding of ribosomes and other proteins — in combination with mass action models and other approaches to disentangle the contributions of the different levels of regulation and generate hypotheses on regulatory mechanisms. We focus on yeast and mammalian cells responding to stress of the endoplasmic reticulum, but have expanded these studies in a variety of directions.