Adjustable Chain Trees for Proteins
Rasmus Fonseca
20 September 2012, 14:00 - 20 September 2012, 15:00 Salle/Bat : 445/PCRI-N
Contact :
Activités de recherche :
Résumé :
A chain tree is a data structure for changing protein conformations. It enables very fast detection of clashes and updates of free energy. A modified version of chain trees that adjust themselves to the changing conformations of folding proteins is introduced. This results in much tighter bounding volume hierarchies and therefore fewer intersection checks. Computational results indicate that the efficiency of the adjustable chain trees is improved by 40% compared to the traditional chain trees. Additionally changing the type of bounding volume from the widely used oriented bounding boxes to a capped cylindrical shape (capsule) decreases the update times by another 40%.
