HAPP: High-accuracy pipeline for processing deep metabarcoding data.

Area of Science:

• Bioinformatics
• Molecular Ecology
• Computational Biology

Background:

• Deep metabarcoding is vital for biodiversity monitoring but faces challenges like noisy data and incomplete databases.
• Accurate diversity estimation and taxonomic annotation are hindered by spurious operational taxonomic units (OTUs) from nuclear-embedded mitochondrial DNA (NUMTs) and sequencing errors.

Purpose of the Study:

• To develop and validate a novel algorithm (NEEAT) and a high-accuracy pipeline (HAPP) for processing deep metabarcoding data.
• To improve the accuracy of diversity estimation and taxonomic annotation in deep metabarcoding studies.
• To benchmark existing tools for chimera removal, taxonomic annotation, and OTU clustering.

Main Methods:

• Introduction of NEEAT algorithm to identify and remove spurious OTUs using 'echo' signals and evolutionary patterns.
• Extensive benchmarking of current tools for chimera removal, taxonomic annotation, and OTU clustering.
• Integration of best-performing tools and parameters into the HAPP pipeline, featuring parallelized computation.

Main Results:

• The HAPP pipeline significantly outperforms existing methods in processing deep metabarcoding data.
• NEEAT effectively removes spurious OTUs originating from NUMTs and sequencing errors.
• HAPP enables efficient analysis of extensive datasets, demonstrated with CO1 data and large-scale insect metabarcoding data.

Conclusions:

• HAPP provides a high-accuracy and efficient solution for deep metabarcoding data analysis.
• The developed methods enhance the reliability of biodiversity monitoring and taxonomic assignment.
• This work addresses key limitations in current deep metabarcoding approaches, paving the way for more robust ecological studies.