Using neural networks to design biosensors
Project duration: January 2023 – June 2025
Midjourney, with prompt: “A protein amino acid sequence transforming into a sensor that gives off fluorescent light. Use a black background.”
Project lead
Dr Andrew Warden
Team
Marltan Wilson, Michael Kuiper, Karine Caron, Cheng Soon Ong, Tom Blau, Janet Reid, Adrian Marsh, Candice Jones, Leigh Gebbie, Lisa Hartman
The challenge
Our ecosystems are under threat from a wide range of chemical contaminants released through human activities such as agriculture, industrial developments, mining, and urbanisation.
Contaminants can have major impacts on aquatic and terrestrial environments, as well as on human health. The ability to detect and monitor contaminants in real-time could help manage and address the negative consequences of contaminants.
One promising approach is to use biosensors, which can specifically detect the presence of a target chemical by converting molecular binding into a measurable signal.
However, designing biosensors that can selectively detect a specific target molecule is a complex task that often requires extensive trial and error experimentation.
Our response
We are using artificial intelligence and machine learning (AIML) methods to more rapidly design effective biosensors. By using AIML, we can rapidly process very large amounts of data to identify patterns that can then be used to generate protein-based biosensor designs.
Designs will be tested for sensitivity and selectivity using high-throughput techniques at the CSIRO BioFoundry. Following successful generation of binding protein candidates, we will partner with the CSIRO Sensors Protoyping Hub to explore device options for device design, catering to the specific application area.
Impact
This approach has the potential to greatly accelerate the development of biosensors for a wide range of applications:
Medical diagnostics: Biosensors can be used to detect biomarkers in body fluids such as blood or urine to diagnose diseases such as cancer, diabetes, and infectious diseases.
Environmental monitoring: Biosensors can be used to detect contaminants such as heavy metals, pesticides, or pathogens in water or soil samples for environmental monitoring.
Food safety: Biosensors can be used to detect toxins, bacteria or other contaminants in food products to ensure food safety and quality.
Agriculture: Biosensors can be used to detect pests or diseases in crops, as well as to monitor soil conditions for optimal plant growth.
Biodefense: Biosensors can be used to detect and identify biological agents such as anthrax or ricin for biodefence and biosecurity purposes.
Drug discovery: Biosensors can be used in drug discovery and development to identify potential drug targets and to screen potential drug compounds for efficacy and toxicity.
Industrial processes: Biosensors can be used to monitor and control industrial processes, such as fermentation or bioreactor systems, for optimal production efficiency and product quality.
Personal health monitoring: Biosensors can be used in wearable devices, such as smartwatches or patches to monitor physiological parameters, glucose or specific biomarker levels, for personal health monitoring and management.