Data-Driven Molecular Design

Developing powerful AI/ML tools to fast-track novel solutions using engineering biology.

Our Data-Driven Molecular Design theme is harnessing rapid advancements in high-throughput data generation, artificial intelligence and machine learning. Our aim is to make innovation in biotechnology orders of magnitude faster, more certain, and more efficient.  

Our approach

The Data-Driven Molecular Design theme draws on CSIRO’s robust capabilities in machine learning, artificial intelligence, quantum biology and computational biology. We make use of world-class national facilities, including the Pawsey Supercomputing Research Centre and the CSIRO BioFoundry

Our researchers build and develop new tools through a learning-by-doing approach. We target specific molecules and bioprocesses, with a view to developing effective workflows then leveraging them for broader applications.

A big part of our process is generating high-quality empirical datasets. These are used to train our AI models for high-throughput and accurate design of biological tools and solutions.   

Example projects include:

  • Building genetic designs for novel microbial production organisms that make food protein, medicines or enzymes that clean up the environment 
  • Designing new biosensors for disease biomarkers, pathogenic microorganisms, or pollutants. 
  • Rapid identification of resistance genes against pathogens in plants, enabling the development of crops that save billions of dollars in food and crop losses each year. 

New tools developed through the DDMD theme will be built through a learning-by-doing paradigm. That is, by targeting specific molecules and bioprocesses with a view to exploiting a much broader application of the workflows once established.

Delivering impact

This theme is producing novel approaches to currently intractable problems, as well as tackle current bottlenecks in biological design. We are focused on developing advanced, enabling technologies and workflows that accelerate scientific discovery and solutions across domains, including health, agriculture, the environment and industry.

We want to bring solutions that were previously thought to be several decades away into a five to ten year window. We are not doing this alone – our collaborators are key to successful delivery of real-world impact. 

By integrating Interdisciplinary Decision-Making with Data-Driven Molecular Design and Innovative Bioproduction Platforms, the Advanced Engineering Biology Future Science Platform will help realise powerful biotechnology solutions for the benefit of Australian society, industry, and environment. 

Contacts

Theme Leader, Data-Driven Molecular Design