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:20241120T082408Z
LOCATION:HG F 26.5
DTSTART;TZID=Europe/Stockholm:20240603T120000
DTEND;TZID=Europe/Stockholm:20240603T123000
UID:submissions.pasc-conference.org_PASC24_sess142_msa172@linklings.com
SUMMARY:Foundational Models and Workflows: Enhancing Deep Learning Compari
 sons in Drug Response Studies
DESCRIPTION:Minisymposium\n\nNeeraj Kumar (Pacific Northwest National Labo
 ratory)\n\nIn the evolving field of computational drug design and discover
 y, the accurate prediction of drug responses through deep learning models 
 remains a significant challenge due to varying methodologies in model impl
 ementation and validation. This inconsistency hampers the objective assess
 ment of model capabilities across different drug representation methods, a
 rchitectures, and datasets. As models become more complex and datasets mor
 e diverse, the necessity for standardized model comparison methodologies b
 ecomes imperative. Traditional comparison approaches, which typically rely
  on performance scores from disparate studies, lead to incomparable and in
 consistent results, obstructing the understanding of factors critical to p
 redictive performance. Addressing this issue, I will discuss our results b
 ased on foundaitonal models and large scale CMP-CV workflow, an automated 
 cross-validation framework designed for the consistent training and evalua
 tion of multiple deep-learning models. By employing standardized datasets,
  preprocessing techniques, and performance metrics, CMP-CV fosters control
 led experimentation while allowing systematic variation in model hyperpara
 meters and architectures. Additionally, the framework supports custom anal
 ytical functions, enabling a more profound investigation into model repres
 entations and associated uncertainties, thereby establishing a more standa
 rdized and comprehensive approach to model comparison in drug response pre
 diction.\n\nDomain: Life Sciences, Computational Methods and Applied Mathe
 matics\n\nSession Chair: Thomas Brettin (Argonne National Laboratory, Univ
 ersity of Chicago)
END:VEVENT
END:VCALENDAR