SuperbugAi Flagship Project

The problem

Antimicrobial resistance (AMR) is one of the greatest threats to human health, with overuse and misuse of antimicrobials in humans and animals accelerating this process.

The rise in AMR imperils the functioning of hospitals and the provision of life-saving treatments such as intensive care, surgery, cancer chemotherapy, organ transplantation and neonatal care. Infections due to AMR pathogens significantly impact patient survival, length of stay and result in billions of dollars of healthcare costs.

The solution

We are now at an exciting crossroads whereby both genomic and digital health technologies can be leveraged using artificial intelligence approaches to develop intelligent ‘learning’ health care systems.

Meet the team

This SuperbugAi flagship project includes an outstanding, multidisciplinary, collaborative team with a combination of expertise and skills that are ideally positioned to succeed in this innovative project. The working groups will maintain momentum in key aspects of the project, and ultimately coordinate their efforts to achieve the goals of the grant.

News

Dr Anton Peleg presented, "SuperbugAI: Integrating electronic medical record and bacterial genomic data to combat antimicrobial resistance", at the HealthData21 conference held in June 2021.

The SuperbugAI project was recently highlighted on the podcast Nights with John Stanley on 2GB. Listen in on the conversation with Dr Anton Peleg, where he talks about the challenges faced in the fight against superbugs as well as the generation of new technology that will see us victorious.

Selected publications

• Macesic, Nenad, et al. Silent spread of mobile colistin resistance gene mcr-9.1 on IncHI2 ‘superplasmids’ in clinical carbapenem-resistant Enterobacterales. Clinical Microbiology and Infection (2021).
• Macesic, Nenad, Oliver J. Bear Don’t Walk IV, Itsik Pe’er, Nicholas P. Tatonetti, Anton Y. Peleg, and Anne-Catrin Uhlemann. Predicting phenotypic polymyxin resistance in Klebsiella pneumoniae through machine learning analysis of genomic data. Msystems 5, no. 3 (2020): e00656-19.
• Chen T, Rigby JD, McGee MD and Tyagi S. Deep learning has potential for harmonising multi-omics data to discover weak regulatory features [version 1; not peer reviewed]. F1000Research 2019, 8:2068 (poster)
• Chen T and Tyagi S. An annotation-free format for representing multimodal data features [version 1; not peer reviewed]. F1000Research 2021, 10(ISCB Comm J):657 (poster)
• Chen T, Abadi AJ, Lê Cao KA and Tyagi S. multiomics: A user-friendly multi-omics data harmonisation R pipeline [version 1; peer review: awaiting peer review]. F1000Research 2021, 10:538