Related Experiment Video
Updated: Dec 7, 2025
07:11
Author Spotlight: Emerging Technologies and Advanced Tools for Decoding Metabolomics Data Analysis
Published on: November 10, 2023
3.0K
A total crapshoot? Evaluating bioinformatic decisions in animal diet metabarcoding analyses
Devon R O'Rourke1,2, Nicholas A Bokulich3, Michelle A Jusino4,5
1Department of Molecular, Cellular, and Biomedical Sciences University of New Hampshire Durham NH USA.
Ecology and Evolution
|October 2, 2020
Summary
Bioinformatic processing significantly impacts animal metabarcoding accuracy. Evaluating denoising, databases, and classification methods is crucial for reliable biodiversity assessments, ensuring results reflect biology, not digital artifacts.
Area of Science:
- Ecology
- Bioinformatics
- Molecular Biology
Background:
- Metabarcoding is vital for assessing biodiversity in mixed communities.
- Standardized bioinformatics practices are lacking for animal diet studies.
- Bioinformatic choices can introduce biases in metabarcoding results.
Purpose of the Study:
- To evaluate the influence of bioinformatic processing on animal metabarcoding outcomes.
- To compare different sequence processing, database, and classification methods.
- To identify best practices for accurate animal metabarcoding analysis.
Main Methods:
- Comparison of denoising versus clustering methods for sequence processing.
- Evaluation of reference databases (GenBank, BOLD) for cytochrome oxidase I (COI) marker gene.
- Assessment of various taxonomic classification tools (BOLD API, vsearch-SINTAX, q2-feature-classifier).
Main Results:
- Denoising methods exhibit lower error rates than clustering, especially after removing low-abundance variants.
- GenBank and BOLD datasets are complementary for COI gene data.
- Taxonomic classification methods significantly affect species identification, with BOLD API showing fewer assignments.
Conclusions:
- Standardized bioinformatics pipelines are essential for robust animal metabarcoding studies.
- Biological mock communities serve as valuable benchmarks for method evaluation.
- Continuous assessment of bioinformatics tools is needed to ensure accuracy and avoid digital artifacts in biodiversity data.
