ساخت اکانت کاربری رایگان

مقالات پذیرفته شده کنگره

  • صفحه نخست
  • مقالات پذیرفته شده اولین کنگره بین المللی ژنتیک سرطان
بازنشر

  • Integrating AI/ML and Multi-Omics Approaches in Pancreatic Cancer Research

  • Armina Barahmand,1,*
    1. Department of medical laboratory , TeMS.C., Islamic Azad University, Tehran, Iran


  • Introduction: Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal malignancies, with survival rates below 10% despite advances in therapy. Its poor prognosis is attributed to an immunosuppressive microenvironment, extensive stromal desmoplasia, and intrinsic resistance to apoptosis. Recent research highlights the potential of integrating multi-omics data with artificial intelligence (AI) and machine learning (ML) methods to identify therapeutic vulnerabilities. This review summarizes a study that investigated TNFRSF10A/TRAILR1 as a novel target in PDAC using a combination of genomics, proteomics, transcriptomics, and computational drug discovery.
  • Methods: The study integrated six independent PDAC genomic datasets, single-cell spatial transcriptomics, and proteomic data. Differentially expressed genes were identified using Bayesian inference and random forest models. Candidate targets were analyzed for mutational frequency, expression patterns, and survival correlations. Regulatory networks involving microRNAs and proteins were constructed to explore post-transcriptional control. Drug discovery incorporated compound screening from FDA-approved and natural libraries, with ADMET filtering, quantitative structure–activity relationship (QSAR) modeling, and transformer-based deep learning (SELFormer). Molecular docking and molecular dynamics simulations validated compound interactions with TNFRSF10A.
  • Results: The analysis identified TNFRSF10A as a significantly upregulated target in PDAC, enriched particularly in carcinoma-associated fibroblast-rich, hypoxic, and immune-suppressive niches. High TNFRSF10A expression correlated with poor survival outcomes. The network analysis revealed interactions with oncogenic miRNAs (miR-182, miR-155, miR-24, and miR-124) and critical signaling molecules. AI-driven screening predicted three promising compounds: temsirolimus, ergotamine, and capivasertib. Docking analysis showed strong binding affinities, particularly for temsirolimus, which maintained structural stability during 300 ns molecular dynamics simulations. These findings suggest that modulating TNFRSF10A may overcome apoptosis resistance and stroma-driven immune exclusion in PDAC.
  • Conclusion: This study demonstrates the value of integrating multi-omics analysis with AI/ML-based drug discovery for PDAC. TNFRSF10A emerged as a novel therapeutic vulnerability enriched in aggressive tumor niches, and drug repurposing strategies identified potential inhibitors. Among these, temsirolimus showed the most stable binding and favorable energetics, highlighting its potential as a TRAILR1 modulator. The findings illustrate a paradigm shift in oncology research, where computational approaches complement biological data to accelerate precision medicine in hard-to-treat cancers.
  • Keywords: Drug Discovery TNFRSF10A/TRAILR1 Pancreatic Cancer Multi_Omics AI/ML

به خانواده بزرگ کنسر ژنتیکس و ژنومیکس سرطان بپیوندید!

ایجاد حساب کاربری رایگان