We are excited to announce our latest publication!
In this study, titled “Harnessing the Power of AI in Precision Medicine: NGS-based Therapeutic Insights for Colorectal Cancer Cohort,” we explore innovative approaches for identifying therapeutic targets in colorectal cancer (CRC) using artificial intelligence.
Methods:
We analyzed data from 71 patients with advanced resectional colorectal adenocarcinoma. Utilizing whole exome sequencing and RNA sequencing, we implemented three key strategies to identify potential therapeutic targets:
Neoepitope Identification: We employed our proprietary neoantigen calling pipeline, ARDentify, in combination with an AI model trained on immunopeptidomics data (ARDisplay) to identify unique neoepitopes in our cohort.
Synthetic Lethality Analysis: By examining recurrent mutations, we selected relevant cancer cell lines and utilized knock-out gene dependency scores to uncover synthetic lethality gene pairs.
AI Model for Patient-Centric Analysis: We developed an AI model that matches patient tumors with representative cell lines using RNAseq and methylation data, facilitating accurate model selection for further validation.
Results:
Our analysis revealed approximately 8,700 unique neoepitopes, with no common targets among more than two patients. We identified three significant synthetic lethality pairs: APC-CTNNB1, BRAF-DUSP4, and the newly noted APC-TCF7L2, which could offer therapeutic implications for patients with APC and BRAF variants. Additionally, we found a potential gene pair, VPS4A and VPS4B, that may hold therapeutic relevance.
Conclusion:
This study highlights the utility of AI methodologies in identifying therapeutic targets for colorectal cancer. Our innovative approaches not only provide valuable insights into our patient cohort but also pave the way for future precision medicine strategies. The development of our AI model for aligning tumors with cell lines marks a significant advancement in tailoring cancer therapies.
Read the full article here.
