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:20230831T095746Z
LOCATION:Hall
DTSTART;TZID=Europe/Stockholm:20230627T193000
DTEND;TZID=Europe/Stockholm:20230627T213000
UID:submissions.pasc-conference.org_PASC23_sess116_pos118@linklings.com
SUMMARY:P57 - Partial Charge Prediction and Pattern Extraction from a Atte
 ntiveFP Graph Neural Network
DESCRIPTION:Poster\n\nMarc Thierry Lehner (ETH Zurich)\n\nMolecular dynami
 cs (MD) simulations enable the time-resolved study of bio-molecular proces
 ses. The quality of MD simulations is, however, highly dependent on the se
 t of interaction parameters used, so-called force fields. The accurate par
 tial-charge assignment of all simulated atoms is hence a crucial part of e
 very MD simulation. Due to the slowly decaying nature of the Coulomb inter
 actions, the effects of different partial-charge assignments can be observ
 ed over long distances and can have drastic effects on the stability of a 
 MD simulation. Therefore, many schemes have been developed over the last d
 ecades to improve partial-charge assignment: Classical tabulated values, a
 b initio calculations, or the prediction with a machine learning model. Ho
 wever, all these approaches have some shortcomings in either accuracy, spe
 ed, or interpretability. Here, we present an option to combine the accurac
 y of ab initio calculations, the speed of machine learning models, and the
  interpretability of tabulated assignments. An attention-based graph neura
 l network is trained on a diverse dataset to predict high-quality atom-in-
 molecule (AIM) partial charges. We then use a model-agnostic approach to e
 xtract the most important sub-graph on an atomistic level to provide the u
 ser with the same level of interpretability as for tabulated values.
END:VEVENT
END:VCALENDAR