From 16S rDNA-seq count data to microbial interactions: A comprehensive benchmark of co-occurrence network inference methods
- 1Department of Information Engineering, University of Padova, Padova, Italy
- 2Department of Comparative Biomedicine and Food Science, University of Padova, Padova, Italy
- 3Padua Center of Network Medicine, University of Padova, Padova, Italy
Abstract
The study of the ecological interactions within microbial communities is crucial for gaining deeper insights into community ensemble and evolution, enabling targeted modification of the community’s functionality. Many inference approaches have been developed to derive ecological associations. Here, we focus on co-occurrence computational methods, in which associations are inferred from metagenomic abundance data by measuring the statistical dependence between species counts across different sampled populations. The limitations in culture-based experiments have hindered the ability to assess these methods’ validity, leading to scarce application in metagenomics studies. To provide a comprehensive assessment of network inference method performance, we benchmark 18 different computational approaches across a broad spectrum of experimental scenarios. We use synthetic 16S rDNA-seq count data generated from known underlying interaction networks, enabling systematic and controlled evaluation. The benchmark spans scenarios varying in numbers of species and samples, sample richness, sequencing depth, and interaction network structure. We evaluated the performance of the methods both in terms of accuracy of the inferred network, using various metrics from binary classification theory and metrics coming from network science, and in terms of computational burden. We found that methods have valid results only in a restricted, and often unrealistic, field of application, whereas their effectiveness diminishes with more realistic simulation parameters. Our findings highlight the limits of the co-occurrence-based approach and the need for further development in microbial ecological association inference. Author summary: The microbiota is the collection of microorganisms that reside within a given ecological environment or niche. Numerous studies have demonstrated that the microbiota is involved in a wide range of biological processes that affect the health or balance of the host organism or ecosystem. Recent advances in high-throughput sequencing technologies, together with progress in bioinformatic and computational techniques, have enabled new discoveries and deeper understanding in this field. One of the most promising analyses in this landscape is the inference of the community interaction network. Researchers are therefore faced with the challenge of selecting appropriate methods from a large and diverse set of available approaches. To address this, we performed a comprehensive benchmark of established co-occurrence microbial network inference tools using simulated data. The use of simulated data was necessary due to the lack of reliable ground-truth references from culture-based experiments, which has limited systematic evaluation of method performance. Our work aims to assess the general applicability of network inference methods and to support researchers in the choice of the best methods available.
Related
- Where single-cell foundation models help: a benchmark of annotation, perturbation and regulatory inference against simple baselines
- Divergent responses of rhizosphere microbial diversity and co-occurrence patterns to elevation and season in Quercus franchetii from the Yuanmou dry‑hot valley, Southwest China
- Navigating the rapidly evolving landscape of sequencing-based spatial transcriptomics technologies: a comprehensive multi-platform benchmark
- How Similar Are Two Brains? A Comprehensive Benchmark of Brain Network Similarity Measures
- Constructing microbiome co-occurrence networks with confidence: A conditional, nonparametric, inference-based approach
BibTeX
@article{mariotto2026from,
title = {{From 16S rDNA-seq count data to microbial interactions: A comprehensive benchmark of co-occurrence network inference methods}},
author = {Piero Mariotto and Matteo Baldan and Barbara Di Camillo and Giacomo Baruzzo},
journal = {PLOS Computational Biology},
year = {2026},
month = oct,
doi = {10.1371/journal.pcbi.1014898},
url = {https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014898},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - From 16S rDNA-seq count data to microbial interactions: A comprehensive benchmark of co-occurrence network inference methods
AU - Piero Mariotto
AU - Matteo Baldan
AU - Barbara Di Camillo
AU - Giacomo Baruzzo
PY - 2026
DA - 2026/10/29
JO - PLOS Computational Biology
DO - 10.1371/journal.pcbi.1014898
UR - https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1014898
AB - The study of the ecological interactions within microbial communities is crucial for gaining deeper insights into community ensemble and evolution, enabling targeted modification of the community’s functionality. Many inference approaches have been developed to derive ecological associations. Here, we focus on co-occurrence computational methods, in which associations are inferred from metagenomic abundance data by measuring the statistical dependence between species counts across different sampled populations. The limitations in culture-based experiments have hindered the ability to assess these methods’ validity, leading to scarce application in metagenomics studies. To provide a comprehensive assessment of network inference method performance, we benchmark 18 different computational approaches across a broad spectrum of experimental scenarios. We use synthetic 16S rDNA-seq count data generated from known underlying interaction networks, enabling systematic and controlled evaluation. The benchmark spans scenarios varying in numbers of species and samples, sample richness, sequencing depth, and interaction network structure. We evaluated the performance of the methods both in terms of accuracy of the inferred network, using various metrics from binary classification theory and metrics coming from network science, and in terms of computational burden. We found that methods have valid results only in a restricted, and often unrealistic, field of application, whereas their effectiveness diminishes with more realistic simulation parameters. Our findings highlight the limits of the co-occurrence-based approach and the need for further development in microbial ecological association inference. Author summary: The microbiota is the collection of microorganisms that reside within a given ecological environment or niche. Numerous studies have demonstrated that the microbiota is involved in a wide range of biological processes that affect the health or balance of the host organism or ecosystem. Recent advances in high-throughput sequencing technologies, together with progress in bioinformatic and computational techniques, have enabled new discoveries and deeper understanding in this field. One of the most promising analyses in this landscape is the inference of the community interaction network. Researchers are therefore faced with the challenge of selecting appropriate methods from a large and diverse set of available approaches. To address this, we performed a comprehensive benchmark of established co-occurrence microbial network inference tools using simulated data. The use of simulated data was necessary due to the lack of reliable ground-truth references from culture-based experiments, which has limited systematic evaluation of method performance. Our work aims to assess the general applicability of network inference methods and to support researchers in the choice of the best methods available.
SN - 1553-7358
ER -