BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Differentiable molecular simulation to improve protein force field
 s - Joe Greener\, MRC Laboratory of Molecular Biology\, Cambridge
DTSTART:20220523T133000Z
DTEND:20220523T140000Z
UID:TALK174641@talks.cam.ac.uk
CONTACT:Chuck Witt
DESCRIPTION:Molecular dynamics has shown success in obtaining biological i
 nsights by providing mechanistic interpretations of experimental data. How
 ever\, the force fields used to describe how the atoms interact are biased
  towards keeping folded proteins folded and fail when applied to disordere
 d proteins or protein aggregation. The idea of using the technique of auto
 matic differentiation\, most commonly used to train neural networks\, to i
 mprove force fields is called differentiable molecular simulation (DMS). I
 t overcomes the limitations of other methods by allowing multiple paramete
 rs to be tuned at once. Prior work has shown that a coarse-grained force f
 ield can be trained from scratch using this approach. Work to correct know
 n deficiencies in all-atom implicit solvent force fields using DMS will be
  described\, along with technical challenges when training such methods.
LOCATION:Venue to be confirmed
END:VEVENT
END:VCALENDAR