PhD defense: Christian Andreetta

Probabilistic equilibrium sampling of protein structures from SAXS data and a coarse-grained Debye formula

Supervision: Assoc. Prof. Thomas Hamelryck

Exam Committee:
Prof. Lise Arleth, Head of studies Physics, X-ray and Neutron Science, University of Copenhagen
Prof. Michael Nilges, Institut Pasteur, Unité de Bioinformatique Structurale
Assoc. Prof. Christian Storm Pedersen, Director of Bioinformatics Center, Department of Computer Science, Aarhus University

Abstract
Small Angle X-ray Scattering (SAXS) is a low-resolution experimental method well suited to study flexible protein complexes in different buffer conditions.

I'll present a calibration protocol for SAXS profiles, and some results of its inclusion in a probabilistic framework for protein structure determination (Phaistos). This protocol identifies a set of maximum-likelihood estimators for the form factors employed in the Debye formula, a theoretical forward model for SAXS profiles. The results compare favourably with state-of-the art tools. Additionally, the algorithm can be directly ported to commodity multicore hardware like Graphic Processor Units (GPUs).

The computation of SAXS profiles, together with Phaistos, has been applied to investigate the clusters of conformations of single-chain, multi-domain proteins in solution.