DPG Phi
Verhandlungen
Verhandlungen
DPG

Berlin 2018 – scientific programme

Parts | Days | Selection | Search | Updates | Downloads | Help

DY: Fachverband Dynamik und Statistische Physik

DY 12: Focus Session: Statistical Physics-Based Methods in Molecular Evolution - organized by Alexander Schug and Martin Weigt (joint session BP/DY)

DY 12.3: Talk

Monday, March 12, 2018, 15:45–16:00, H 2013

Coevolution based inference of allosteric architectures — •Barbara Bravi1, Carolina Brito2, Riccardo Ravasio1, and Matthieu Wyart11Institute of Theoretical Physics, Ecole Polytechnique Fédérale de Lausanne, Switzerland — 2Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil

We analyze maximum entropy approaches to infer the functional design of elastic materials exhibiting allostery, i.e. the property of highly specific responses to ligand binding at a distant active site. To guide and inform protocols of de novo drug design, it is fundamental to understand what architectures underlie such a transmission of information and whether their features can be predicted from sequence data alone. We consider the functional designs of in silico evolved allosteric architectures which propagate efficiently energy (including shear, hinge, twist) or strain (resulting in a less-constrained trumpet-shaped region between the allosteric and the active site). We show that maximum entropy approaches, built to capture statistical properties such as conservation and correlations, can provide predictive information on the cost of single and double mutations while their performance at reproducing the original allosteric fitness is strongly design-dependent. We benchmark existing maximum entropy inference methods on these computationally evolved functional architectures and we propose an improved framework accounting for a multiplicity of co-evolutionary factors which is aimed at disentagling allostery-based correlations from extrinsic ones.

100% | Mobile Layout | Deutsche Version | Contact/Imprint/Privacy
DPG-Physik > DPG-Verhandlungen > 2018 > Berlin