BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260202T201803Z
LOCATION:241
DTSTART;TZID=America/Chicago:20251117T115500
DTEND;TZID=America/Chicago:20251117T121000
UID:submissions.supercomputing.org_SC25_sess205_ws_cafcws104@linklings.com
SUMMARY:PathPCNet: Pathway Principal Component-Based Interpretable Framewo
 rk for Drug Sensitivity Prediction
DESCRIPTION:Bikhyat Adhikari, Masrur Sobhan, Ananda Sutradhar, Giri Narasi
 mhan, and Ananda Mohan Mondal (Florida International University)\n\nBackgr
 ound: Precision medicine aims to identify significant biomarkers and effec
 tive drugs based on individual genomic profiles, enabling personalized tre
 atment strategies. Drug efficacy is commonly assessed via drug response, t
 ypically measured by the concentration required to inhibit a biological ac
 tivity (e.g., IC50). In contrast, drug sensitivity reflects the strength o
 f a tumor's response to a drug, where a lower effective dose indicates hig
 her sensitivity. With the increased availability of large-scale multi-omic
 s datasets, machine learning (ML) and deep learning approaches have emerge
 d as powerful tools for studying drug response--holding great promise for 
 accelerating biomarker discovery and enabling the development of more effe
 ctive therapeutics.\n\nMethods: We present `PathPCNet`, a novel interpreta
 ble deep learning framework that integrates multi-omics data (copy number 
 variation, mutation, and RNA sequencing) with biological pathways, drug mo
 lecular structures, and Principal Component Analysis (PCA) to predict drug
  response. We project high-dimensional, noisy gene-level features to pathw
 ay-level principal components, and evaluate six machine learning models us
 ing the first one to five principal components. Our models are trained to 
 predict the IC50 values for 182 drugs across 409 cell lines representing 2
 9 cancer types from the GDSC (Genomics of Drug Sensitivity in Cancer) data
 set. Finally, we fine-tune the deep learning model and apply SHAP to inter
 pret feature contributions. SHAP scores are back-projected from the princi
 pal components to original genes using PCA loadings, enabling identificati
 on of the most significant genes.\n\nResults: Our model achieves a Pearson
  correlation coefficient of 0.941 and an R-squared value of 0.885, outperf
 orming existing pathway-based approaches for drug response prediction. Usi
 ng SHAP-based model interpretation, we quantify the contributions of diffe
 rent omics and drug features, and identify critical pathways and gene-drug
  interactions involved in resistance mechanisms. These results highlight t
 he potential of integrative deep learning models not only for accurate pre
 diction, but also for uncovering biologically meaningful insights that can
  inform drug discovery and precision oncology. Furthermore, our framework 
 enables the identification of key pathways, genes, and atomic-level drug a
 ttributes associated with drug sensitivity across diverse cancer types.\n\
 nDiscussion: Our intuitive feature extraction approach, based on pathway-l
 evel principal components, effectively reduces dimensionality while preser
 ving data variance and enhancing biological interpretability. Tumor respon
 se is a complex biological phenomenon that extends beyond single gene–drug
  interactions. Therefore, integrating multi-omics profiles and molecular d
 rug features within the context of biological pathways is essential for un
 derstanding drug response. This integrative approach has strong potential 
 to support targeted therapy design, biomarker discovery, and the advanceme
 nt of precision medicine and drug development.\n\nRecording: Livestreamed,
  Recorded\n\nRegistration Category: Technical Program Reg Pass, Workshop R
 eg Pass\n\nSession Chairs: Eric Stahlberg (MD Anderson Cancer Center, Univ
 ersity of Texas); Sally Ellingson (University of Kentucky); Lynn Borkon (F
 rederick National Laboratory for Cancer Research); Patricia Kovatch (Icahn
  School of Medicine at Mount Sinai); Lauren Lewis (Frederick National Labo
 ratory for Cancer Research); and Sean Hanlon (National Institutes of Healt
 h (NIH), National Cancer Institute (NCI))\n\n
END:VEVENT
END:VCALENDAR