BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:Europe/Stockholm
X-LIC-LOCATION:Europe/Stockholm
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230831T095745Z
LOCATION:Schwarzhorn
DTSTART;TZID=Europe/Stockholm:20230626T143000
DTEND;TZID=Europe/Stockholm:20230626T150000
UID:submissions.pasc-conference.org_PASC23_sess167_msa182@linklings.com
SUMMARY:Enhancing Molecular Dynamics Simulations through Deep Learning
DESCRIPTION:Minisymposium\n\nFelix Pultar, Moritz Thürlemann, and Sereina 
 Riniker (ETH Zurich)\n\nComputational chemistry allows exploration of mole
 cular systems at atomic resolution. Molecular dynamics (MD) simulations ar
 e key to account for solvent and entropic effects. If an explicit descript
 ion of the system's electronic structure is also required, mixed quantum m
 echanics/molecular mechanics (QM/MM) MD simulations are highly attractive.
  However, application of these methods is limited due to their computation
 al cost. Semi-empirical methods like PM7, DFTB, or XTB approximate the Ham
 iltonian, which leads to significant reduction of computational costs. To 
 alleviate the inherent loss in accuracy, we have developed deep neural net
 works that predict the difference in the potential-energy surface of the Q
 M particles calculated with DFTB compared to a DFT reference (Δ-learning).
  Newer generations of these models are also designed to approximate the re
 ference method itself. We have explored high-dimensional neural network po
 tentials, as well as graph-convolutional neural network architectures. Mor
 e recently, we demonstrated the possibility of encoding long-distance and 
 directional information by means of an anisotropic message passing formali
 sm. The deep-neural networks are interfaced to the GROMOS software to allo
 w condensed-phase (QM)ML/MM MD simulations at DFT level of theory at a fra
 ction of the attendant computational costs. Large systems that were hither
 to unfeasible to explore will thus become available for in-depth study.\n\
 nDomain: Chemistry and Materials\n\nSession Chair: Julija Zavadlav (Techni
 cal University of Munich)
END:VEVENT
END:VCALENDAR