Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
April 5, 2026Cancer Research

Comprehensive LLM-Enabled Pharmacodynamic Biomarker Resource for Cancer Drug Development

View Full Paper
Ask AI
Bookmark
Share

Authors

YYYuntao YangLZLi ZhaoSOSeyedmehdi Orouji

Discussion

Loading...

Member takes

Overview

This framework identifies target-specific biomarkers in cancer, suggesting improved drug development outcomes.

Key Points

  • The aim is to develop a dataset that connects drug targets to validated pharmacodynamic biomarkers for cancer therapeutics.
  • Curated biomarker candidates from genomic and pharmacologic resources across nine target classes.
  • Cross-referenced multiple databases to refine protein classifications and improve annotation accuracy.
  • Utilized canSAR interactome to identify target-biomarker interactions supported by experimental evidence.
  • Computed cohort-specific correlations using TCGA, TARGET, and GTEx datasets.
  • Employed an LLM-based fact-checking agent to harmonize antibody annotations focused on enzyme targets.
  • Proposed 73,000 high-confidence target-biomarker relationships involving over 2,100 potential drug targets.
  • 67% of the relationships are transcription factor-gene interactions, while 33% are enzyme-substrate interactions.
  • Identified commercial antibodies for over 2,800 biomarker candidates for experimental validation.
  • The dataset covers more than 60% of the top 20 predicted targets across 19 cancer types.

Cite This Study

Yang et al. (2026) studied this question.

synapsesocial.com/papers/69d1fdd4a79560c99a0a41a6https://doi.org/10.1158/1538-7445.am2026-2720
View Full Paper
Ask AI
Bookmark
Share