PhD opportunities – AI-driven Whole Cell Modelling (MACSYS @ ANU)

We invite applications for two fully funded PhD positions (3.5 years) based at the ARC Centre of Excellence in Mathematical Analysis of Cellular Systems (MACSYS) node at The Australian National University (ANU). These projects sit at the interface of machine learning, mathematical modelling, and computational biology, aiming to push the frontier of AI-enabled whole-cell and RNA-focused modelling.

2026 PhD application rounds: April (domestic & international), August (international), October (domestic)

To apply / express interest: Please email the supervisor(s) with

1. a brief cover note (email is fine) outlining your background + which project(s) you’re interested in,

2. CV, and

3. academic transcripts.

Project 1 (AI-driven Whole Cell Modelling)

We are interested in these research areas: (1) Hybrid Whole Cell Modelling: Combine mechanistic cell models with modern ML to deliver robust, well-calibrated predictions across conditions and perturbations. (2) Agentic AI Virtual Cell: Integrate molecular foundation models with single-cell multi-omics data and use agentic/active learning to propose the next best perturbations to improve generalisation. (3) Genomic & RNA Foundation Models: Train genome/RNA foundation models that translate sequence and regulation to function and phenotype, enabling principled in silico perturbations and design. (4) Multi-omics Modelling: Fuse transcriptomics, proteomics, metabolomics and perturbation data to infer cell states and dynamics that power whole-cell model construction.

Students with backgrounds in computer science, mathematics, or computational biology and strong ML/mathematical modelling skills are encouraged to apply. 

Supervisor contact: Associate Professor Jiayu (Jean) Wen  (Jiayu.wen@anu.edu.au)

Project 2 (AI-driven decoding and design of RNA-interacting molecules)

AI-Driven Decoding and Design of RNA-interacting molecules. This PhD project will develop integrated AI systems for biomedicine to decode and engineer molecules that specifically target RNA, a key challenge for cell biology, biotechnology, and RNA-targeting therapeutics. It aims to explore the combination of RNA-aware deep learning models, diffusion-based molecular generation, and agentic reasoning to design synthetic RNA-interacting molecules. The research leverages multimodal training on sequences, structures, and interaction data (including RNA chemical modifications), incorporates domain-specific biological constraints, and explores agentic AI to build hypothesis-driven workflows that bridge interaction prediction, rational molecular design, and experimental validation, thereby accelerating the development of programmable RNA-targeting tools.

Supervisor contact: Professor Eduardo Eyras (Eduardo.eyras@anu.edu.au)