European Commission logo
English English
CORDIS - EU research results

Machine learning based analytics for bacteria cell cycle characterization using super resolution microscopy

Project description

Bacterial cell cycle characterisation using super-resolution microscopy

The cell cycle (CC) in bacteria involves an internal clock associated, in particular, with cell size homeostasis at a population level. Some analytical tools dedicated to characterising the CC are available for eukaryotic cells, yet they are inadequate when studying bacterial cells. The rise of antibiotic resistance requires reliable quantitative platforms dedicated to the bacterial CC that allow for high-throughput analysis. The EU-funded BALTIC project aims to develop a novel approach for bacterial CC investigation. The methodology will rely on cutting-edge high-throughput super-resolution microscopy to construct dynamic models of the bacterial CC using fixed cells. The technology represents a considerable step forward, benefitting from a spatial resolution of 10 nm and bacterial cell quantitative information.


The cellular life cycle, or cell cycle (CC), is the fundamental backbone of the cellular machinery: it orchestrates processes over multiple scales, in space and time. In bacteria, it consists of an internal clock associated with, for instance, cell-size homeostasis at a population level. Although some analytical tools dedicated to characterizing CC are available for eukaryotic cells, such approaches are still lacking when it comes to bacteria cells study. Moreover, existing eukaryotic cell cyclers are highly limited in terms of both resolution (spatial or temporal) and applications. However, with the rise of antibiotic resistance, there is a real need for reliable quantitative platforms dedicated to bacteria CC and allowing for high throughput comparison studies. This research proposal aims at producing a novel approach for bacteria CC investigation, whilst over-passing the drawbacks associated with existing tools. The developed methodology will rely on cutting edge high throughput super resolution microscopy. We will firstly explore proteins contribution to characterizing CC at the nano-scale, taking a step back from the unreliable and limited size or time dependent estimation. Relying on state of the art machine learning strategies and the identified CC reporting features, I will develop tactics to circumvent the trade-off between temporal and spatial resolution constraining fluorescence nanoscopy when it comes to the study of dynamic processes such as CC. I will implement a methodology to extract, for the first time, dynamic models of bacteria CC from fixed cells super resolved images. It is a considerable step forward: enabling to benefit from a spatial resolution around 10 nm, whilst inferring live-cell akin quantitative information. The highly innovative approaches to bacteria CC quantification developed here will be made generalizable across cell types and applications, providing a unique platform for complex studies, and therapeutics development.


Net EU contribution
€ 191 149,44
1015 Lausanne

See on map

Schweiz/Suisse/Svizzera Région lémanique Vaud
Activity type
Higher or Secondary Education Establishments
Total cost
€ 191 149,44