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:20250822T115805Z
LOCATION:Campussaal - Plenary Room
DTSTART;TZID=Europe/Stockholm:20250617T103000
DTEND;TZID=Europe/Stockholm:20250617T110000
UID:submissions.pasc-conference.org_PASC25_sess150_pos156@linklings.com
SUMMARY:P30 - The MENTOR Interpretation Agent: From Network Embeddings to 
 Mechanistic Narratives via Retrieval-Augmented LLMs
DESCRIPTION:Anna H.C. Vlot (Oak Ridge National Laboratory); Matthew Lane (
 Oak Ridge National Laboratory; Bredesen Center for Interdisciplinary Gradu
 ate Research and Education, University of Tennessee-Knoxville); Kyle A. Su
 llivan (Oak Ridge National Laboratory); Peter Kruse (Oak Ridge National La
 boratory; Bredesen Center for Interdisciplinary Graduate Research and Educ
 ation, University of Tennessee-Knoxville); John Dandy and Selin Kaplanoglu
  (Oak Ridge National Laboratory); Alice Townsend and Jean Merlet (Oak Ridg
 e National Laboratory; Bredesen Center for Interdisciplinary Graduate Rese
 arch and Education, University of Tennessee-Knoxville); and Daniel A. Jaco
 bson (Oak Ridge National Laboratory)\n\nDespite an increasing number of co
 mplex omics data sets, extracting comprehensive mechanistic insights from 
 these data remains challenging. To address this, we developed a human-in-t
 he-loop LLM-based agentic retrieval-augmented generation (RAG) pipeline, t
 he MENTOR Interpretation Agent (MENTOR-IA), to identify novel relationship
 s among multi-omic gene sets. We applied MENTOR-IA to interpret a previous
 ly characterized set of 211 opioid addiction-related genes. We first parti
 tioned these genes into clades using hierarchical clustering of random wal
 k with restart (RWR)-based graph embeddings presented in a dendrogram usin
 g our previously described MENTOR algorithm. MENTOR-IA identified Akt, ERK
 , and BDNF signaling pathways known to be critical to synaptic plasticity,
  previously reported to be associated with the 211 opioid addiction-relate
 d genes. In addition, our pipeline identified novel biological processes l
 ike extracellular matrix remodeling and vasculogenesis that were not ident
 ified through prior manual review. These results illustrate that our integ
 rative pipeline facilitates scalable interpretation of multi-omic datasets
 , accelerating our capability to comprehend complex biological traits. Ult
 imately, these innovations will enhance our ability to derive actionable i
 nsights for disease biology and therapeutic development from multi-omic da
 ta.\n\nSession Chair: David Moxey (King's College London)\n\n
END:VEVENT
END:VCALENDAR