Bayesian semi-supervised classification of bacterial samples using MLST databases
2011

Bayesian Classification of Bacterial Samples

Sample size: 515 publication Evidence: high

Author Information

Author(s): Cheng Lu, Connor Thomas R, David M Aanensen, Brian G Spratt, Jukka Corander

Primary Institution: University of Helsinki

Hypothesis

Can a Bayesian model-based method improve the classification of bacterial samples using MLST databases?

Conclusion

The study introduces a tool for automated semi-supervised classification of new pathogen samples, enhancing the analysis of bacterial strains in relation to existing databases.

Supporting Evidence

  • The method allows for the classification of new strains into known and novel groups.
  • It provides probabilistic quantification of classification uncertainty.
  • The tool is computationally efficient, enabling rapid analysis of large datasets.

Takeaway

This study created a smart tool that helps scientists quickly figure out where new bacteria samples fit in with known ones, making it easier to study diseases.

Methodology

The study used a Bayesian model-based method for semi-supervised classification of MLST data, allowing for the classification of new bacterial isolates against existing databases.

Limitations

The method's effectiveness may vary based on the quality and completeness of the MLST databases used.

Digital Object Identifier (DOI)

10.1186/1471-2105-12-302

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication