BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160552Z
LOCATION:Track 3
DTSTART;TZID=America/New_York:20201113T130000
DTEND;TZID=America/New_York:20201113T131500
UID:submissions.supercomputing.org_SC20_sess222_ws_cafcw125@linklings.com
SUMMARY:Deep Learning Based Prediction of the Temporal Behavior of RAS Pro
 tein Conformations on Simulated Cell Membrane Surfaces
DESCRIPTION:Workshop\n\nDeep Learning Based Prediction of the Temporal Beh
 avior of RAS Protein Conformations on Simulated Cell Membrane Surfaces\n\n
 Moody, Bhatia, Bremer, Carpenter, Dharuman...\n\nApproximately 30% of huma
 n cancers are caused by interactions of a mutated RAS protein with the dow
 nstream RAF near cell membranes.  The DoE Cancer Pilot 2 campaign conducte
 d numerous molecular dynamics (MD) simulations modeling RAS proteins in co
 ntact with the lipid bilayer of a cell membrane.  In this work, we present
  a new deep learning (DL) based technique to explore temporal correlations
  between the protein and the 14 types of lipid species. Our early experime
 nts have already led to several key findings.\n\nWe investigate the applic
 ation of transformer models, such as BERT, which train a deep neural netwo
 rk to predict the past and future changes in RAS conformations given a seq
 uence of simulation frames as input.  Given an ordered sequence of frames 
 of the lipid concentrations around the RAS protein, the models identify pe
 riods where the RAS conformation is stable or likely to change state with 
 60-70% accuracy.  Models trained on sequences of frames of the distance of
  the RAS from the cell membrane achieve accuracy of 85-90%.   We show that
  model accuracy improves with increasing sequence length and that the mode
 ls utilize time-based information in the input sequences.\n\nOur initial e
 xperiments indicate the possibility of extracting temporal correlations fr
 om MD simulations using appropriate DL models. Our work paves way to a new
  form of analysis for MD trajectories focusing on the prediction of events
  of interest. There still remain many open questions, foremost of which pe
 rtains to distilling the correlations to understand the order of events an
 d assess causation.\n\nRegistration Category: Workshop Reg Pass
END:VEVENT
END:VCALENDAR