Statistical Inference for Multi-Pathogen Systems
2011

Detecting Interactions Among Pathogens Using Likelihood-Based Inference

publication Evidence: high

Author Information

Author(s): Shrestha Sourya, King Aaron A., Rohani Pejman

Primary Institution: University of Michigan

Hypothesis

Can a likelihood-based inference framework accurately detect and quantify the presence and nature of pathogen interactions from epidemiological data?

Conclusion

The study demonstrates that likelihood-based inference can effectively identify and quantify interactions between pathogens, even in the presence of noise and stochasticity.

Supporting Evidence

  • Likelihood-based inference can accurately detect pathogen interactions even with noisy data.
  • Stronger and longer-lasting interactions are more easily quantified.
  • Phase relationships alone are unreliable indicators of pathogen interactions.

Takeaway

This study shows that scientists can use data from sick people to figure out how different germs interact with each other, which helps in understanding diseases better.

Methodology

The study used a likelihood-based inference framework to analyze simulated epidemiological data for detecting pathogen interactions.

Potential Biases

Potential biases from under-reporting and aggregation of data were considered, but the method showed robustness against these issues.

Limitations

The approach may be limited by assumptions of known epidemiological parameters and the quality of the data used.

Digital Object Identifier (DOI)

10.1371/journal.pcbi.1002135

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication