Research Programs

The AIEDD CRC proposes three interconnected and synergistic research programs (RPs) designed to drive innovation. The RPs will adopt a comprehensive patient-centric framework to drug discovery, integrating patient data and outcomes to guide more informed and clinically relevant decision making throughout the development process.

Underpinning, supporting and enabling the 3 RPs are three cross-cutting CRC Pillars, bringing together expertise and resources in (Pillar A) education and training; (Pillar B) data, technology and computing infrastructure; and (Pillar C) commercialisation and clinical translation.

RP1 will:

Integrate diverse -omics and non-omics datasets into platforms that can be used to identify more translatable drug targets.

Develop foundational AI models to better explain disease heterogeneity and improve target selection.

Develop a national coordinated framework for the collection, storage and analysis of human biological samples and associated data.

Identify new molecular probes and corresponding biomarkers to interrogate clinically relevant pathways and drug targets.

Apply new AI/ML tools to stratify clinical cohorts of patients based on the identification of common biological pathways and targets.

RP2 will:

Develop and use human-originated, disease-relevant in vitro models (e.g. 3D cell cultures, organoids, cells-on-a-chip platforms) for high throughput validation of novel targets.

Develop AI tools to extract rich phenotypic information from in vitro models to build phenotypic maps of complex disease biology.

Pilot new computational approaches (e.g. digital twins) to better predict disease-site responses to drugs in silico, including pharmacokinetic and pharmacodynamic considerations.

Use AI to develop validation methods such as assays, e.g., using the molecular probes and biomarkers.

Co-develop validation frameworks for new approach methodologies with relevant regulatory bodies (e.g. TGA) to ensure models and readouts represent acceptable evidence for drug approval.

RP3 will:

Use AI-enhanced technologies to propose and prioritise molecular entities against identified drug targets (including from RP1), and refine these models incorporating feedback loops from drug candidate testing.

Synthesise and test highly-ranked drug candidates using human-derived in vitro models and diagnostic assays (including those developed in RP2).

Build aggregate in silico models to predict multi-system ADME/PK/PD behaviour and on- and off-target responses, enabling system-level optimisation of candidates.

Develop AI/ML tools that map molecular interaction profiles with cellular and tissue-level phenotypes and adverse events to anticipate off-target interactions and toxicity.

Leverage existing multi-modal datasets to train prediction tools and probe clinical datasets for drug repurposing opportunities that can be tested against RP1 targets in RP2 models.