Standardizing food web reconstruction from metabarcoding data using iterative rarefaction and trait-based trophic...

Area of Science:

• Ecology
• Molecular Ecology
• Bioinformatics

Background:

• High-throughput metabarcoding and trait-based trophic inference can reconstruct food webs from environmental DNA (eDNA).
• Uneven sequencing depth in metabarcoding data can introduce significant bias into inferred food web structures and network metrics.
• Current methods often rely on unstandardized read counts, potentially conflating sampling effort with ecological processes.

Purpose of the Study:

• To develop an objective and reproducible framework for standardizing food web metrics derived from metabarcoding data.
• To mitigate biases caused by uneven sequencing depth in amplicon-based food web reconstructions.
• To enable more reliable comparisons of food web structures across different environmental conditions.

Main Methods:

• Integration of trait-based trophic inference with iterative rarefaction of metabarcoding community data.
• Development of rarefaction curves based on food web network properties, rather than traditional taxonomic richness.
• Determination of optimal rarefaction thresholds using a first derivative criterion for robust metric estimation.

Main Results:

• The proposed framework successfully reduced sequencing-depth bias in soil protist food web analysis.
• Environmental drivers influencing food web structure became apparent after standardization, which were masked in non-rarefied analyses.
• The method provides reliable estimates of food web structure that are independent of sequencing effort.

Conclusions:

• This iterative rarefaction framework offers a standardized solution for analyzing amplicon-derived food webs.
• The approach facilitates robust comparisons of food web structures across environmental gradients.
• The method enhances the ecological interpretability of metabarcoding data by minimizing technical artifacts.