Statistical approaches to account for false-positive errors in environmental DNA samples.

Area of Science:

• Ecology
• Molecular Biology
• Statistical Modeling

Background:

• Environmental DNA (eDNA) sampling is a powerful tool for biodiversity assessment.
• eDNA analysis is susceptible to both false-positive and false-negative detection errors.
• Inaccurate error handling can compromise ecological inference from eDNA data.

Purpose of the Study:

• To review and compare statistical methods for accounting for errors in eDNA data analysis.
• To evaluate the impact of false positives and negatives on occupancy and detection estimates.
• To provide practical guidance and code for implementing robust eDNA analysis methods.

Main Methods:

• Statistical method review for eDNA error analysis.
• Simulation studies to compare different modeling approaches.
• Exploration of ad hoc versus evidence-based methods for handling detection errors.

Main Results:

• Even low false-positive rates significantly bias occupancy and detectability estimates.
• Ad hoc removal of single PCR detections leads to biased results.
• Alternative methods using prior information or ancillary data improve accuracy.

Conclusions:

• Routinely adopting robust statistical methods is essential for accurate eDNA studies.
• Careful consideration of false positives and negatives is critical for reliable ecological conclusions.
• Advanced modeling approaches mitigate biases inherent in standard eDNA sampling.