BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Two for One: Diffusion Models and Force Fields for Coarse-Grained 
 Molecular Dynamics - Marloes Arts\, University of Copenhagen\, Denmark
DTSTART:20231023T130000Z
DTEND:20231023T133000Z
UID:TALK207499@talks.cam.ac.uk
CONTACT:Eszter Varga-Umbrich
DESCRIPTION:Coarse-grained (CG) molecular dynamics enables the study of bi
 ological processes at temporal and spatial scales that would be intractabl
 e at an atomistic resolution. However\, accurately learning a CG force fie
 ld remains a challenge. In this work\, we leverage connections between sco
 re-based generative models\, force fields\, and molecular dynamics to lear
 n a CG force field without requiring any force inputs during training. Spe
 cifically\, we train a diffusion generative model on protein structures fr
 om molecular dynamics simulations\, and we show that its score function ap
 proximates a force field that can directly be used to simulate CG molecula
 r dynamics. While having a vastly simplified training setup compared to pr
 evious work\, we demonstrate that our approach leads to improved performan
 ce across several protein simulations for systems up to 56 amino acids\, r
 eproducing the CG equilibrium distribution and preserving the dynamics of 
 all-atom simulations such as protein folding events.
LOCATION:Zoom link: https://zoom.us/j/92447982065?pwd=RkhaYkM5VTZPZ3pYSHpt
 UXlRSkppQT09
END:VEVENT
END:VCALENDAR