From 16S rDNA-seq count data to microbial interactions: A comprehensive benchmark of co-occurrence network inference methods.

Benchmarking 18 co-occurrence network inference methods on synthetic 16S rDNA-seq data shows valid results only in a restricted, often unrealistic, application range.

• Why it matters: Culture-based validation is limited, leaving metagenomic co-occurrence methods hard to assess, and their reliability under realistic sampling conditions remains unclear.
• What they did: Synthetic 16S rDNA-seq count data from known interaction networks were used to test methods across varying species and sample counts, richness, sequencing depth, and network structure.
• The result: Performance metrics for inferred networks and computational burden indicate effectiveness drops as simulation parameters become more realistic, highlighting limits of co-occurrence-based association inference.

AI summary of the abstract · may contain errors · not medical advice · read the paper.