AI-driven drug discovery

Our research team pioneered the development of GIPCi. This is a first-in-class small-molecule inhibitor targeting the PDZ domain of GAIP-interacting protein C-terminus 1 (GIPC1), a critical scaffolding protein overexpressed in pancreatic ductal adenocarcinoma (PDAC). Using advanced AI-powered generative design, molecular modeling, and hydrogen-deuterium exchange mass spectrometry (HDX-MS) validation, we successfully identified and optimized GIPCi to disrupt GIPC1-mediated oncogenic signaling.

Preclinical studies demonstrated that GIPCi significantly suppressed tumor growth, enhanced the efficacy of gemcitabine, remodeled the tumor microenvironment, and prolonged survival in multiple PDAC models — all without detectable toxicity. Transcriptomic and spatial transcriptomic analyses revealed immune activation, extracellular matrix remodeling, and downregulation of key tumor-promoting pathways, highlighting its potential for combination therapies.

This work exemplifies how AI-assisted drug discovery can unlock previously undruggable targets, accelerating the translation of innovative therapies for treatment-resistant cancers.