High-throughput bacterial aggregation analysis in droplets.

Microfluidic droplet platforms provide a rapid tool to study and capture bacterial aggregation in a well-controlled micro-environment, while image analysis presents an easily available technique to investigate droplet contents. However, the lack of standardised, well-documented methods and reliance on custom image analysis workflows limits wider adoption of the method and produces inconsistent, incomparable data on aggregation. We present a robust, cost-effective method using both mono- and polydisperse droplets and texture-based image analysis via an open-source software CellProfiler™ to assess bacterial aggregation. Compared to a manual droplet evaluation carried out by a human expert panel, textural characterisation achieves accuracy over 90% and more than 80% precision. Applying the pipeline, we found that suboptimal antibiotic concentrations can increase aggregation, whereas exposure to microplastic beads and metals reduces it. Overall, the developed pipeline offers high accuracy, easy setup, and broad applicability for bacterial aggregation.