OutbreakFinder: a visualization tool for rapid detection of bacterial strain clusters based on optimized...

Area of Science:

• Microbiology
• Computational Biology
• Epidemiology

Background:

• Next-generation sequencing (NGS) enables whole-genome sequencing of bacterial isolates for epidemiological studies.
• Phylogenetic analysis is commonly used to assess bacterial relatedness and identify potential outbreaks.
• Interpreting complex phylogenetic trees requires specialized evolutionary biology knowledge, posing a barrier for many researchers.

Purpose of the Study:

• To develop an accessible tool, OutbreakFinder, that simplifies the interpretation of phylogenetic data for epidemiological investigations.
• To provide a user-friendly interface for analyzing bacterial isolate relatedness and detecting potential outbreaks.

Main Methods:

• OutbreakFinder accepts various input formats: distance matrices (CSV), alignment files (Lyve-SET, Parsnp, ClustalOmega), and Newick tree files.
• The tool employs multidimensional scaling (MDS) for dimension reduction.
• Affinity propagation clustering is utilized to identify clusters within the reduced data, generating a 2D plot.

Main Results:

• OutbreakFinder generates a cluster-labeled two-dimensional plot from phylogenetic data.
• The tool facilitates easier visualization and interpretation of epidemiological relatedness among bacterial isolates.

Conclusions:

• OutbreakFinder offers a simplified approach to analyzing bacterial whole-genome sequencing data for epidemiological purposes.
• The tool democratizes the use of phylogenetic analysis in outbreak detection, assisting researchers without extensive evolutionary biology training.