Integrating Bayesian Analysis and Cumulative Probability Generates High Confidence Using a Single Microbial Source...

Area of Science:

• Environmental microbiology
• Molecular biology
• Water quality assessment

Background:

• Microbial source tracking (MST) is crucial for identifying fecal contamination in waterbodies.
• Existing MST assays lack both perfect host specificity and sensitivity for human or animal waste.
• Host-associated molecular markers are key tools in MST.

Purpose of the Study:

• To apply Bayes' Theorem to assess the reliability of the HF183 human fecal-associated marker in MST.
• To evaluate the impact of marker specificity and sensitivity on contamination detection.
• To determine the confidence levels achievable with concurrent marker use and multiple sampling.

Main Methods:

• Bayes' Theorem was used with literature-derived sensitivity and specificity values for the HF183 marker.
• Type I and Type II error rates were analyzed across various prior probabilities.
• Conditional probabilities were calculated for concurrent use of HF183 and HumM2 markers.
• Cumulative probability analysis was performed using multiple samples to assess detection likelihood.

Main Results:

• Marker specificity and sensitivity values directly influence the accuracy of true positive and true negative results.
• Concurrent use of HF183 and HumM2 markers, with practical priors, achieved 99% confidence in results.
• Cumulative probability analysis demonstrated that multiple samples (>95% confidence) can yield reliable positive and negative results.

Conclusions:

• Bayes' Theorem provides a robust framework for improving the reliability of MST.
• Concurrent use of multiple human fecal markers enhances confidence in identifying contamination sources.
• Multiple sampling events at a single site can compensate for in situ variability and ensure accurate MST results, even with a single marker.