VTAM: A robust pipeline for validating metabarcoding data using controls.

Area of Science:

• Ecology
• Bioinformatics
• Molecular Biology

Background:

• Metabarcoding studies require careful design to minimize false positives (FP) and false negatives (FN) for accurate ecological estimates.
• Existing metabarcoding software often fails to integrate experimental data like internal controls and replicates for optimal error reduction.

Purpose of the Study:

• To develop and evaluate VTAM, a novel metabarcoding analysis software designed to explicitly use experimental design elements for optimizing parameter settings.
• To minimize FP and FN occurrences in metabarcoding data through a stringent filtering procedure.

Main Methods:

• Developed VTAM, a Python-based software that integrates internal controls, replicates, and overlapping markers.
• Benchmarked VTAM against two other pipelines using three datasets and two markers (COI, 16S).
• Implemented a stringent filtering procedure within VTAM to enhance data robustness.

Main Results:

• VTAM demonstrated comparable sensitivity to existing pipelines but with significantly higher precision.
• The software effectively utilizes experimental design data to optimize filtering parameters.
• VTAM's approach leads to more robust metabarcoding data.

Conclusions:

• VTAM offers a non-arbitrary and standardized method for validating metabarcoding data.
• The software is crucial for obtaining accurate ecological estimates in biodiversity studies.
• VTAM represents a significant advancement in reliable metabarcoding data analysis.