Comparer des méthodes

	Analyse bayésienne de la diversité du microbiome	Analyse de la diversité du microbiome basée sur les réseaux
Domaine	Bio-informatique	Bio-informatique
Famille	Process / pipeline	Process / pipeline
Année d'origine	2010s (Dirichlet-Multinomial approach formalized ~2012; extensions ongoing)	2012
Auteur d'origine	Ian Holmes, Katie Harris, Christopher Quince (Dirichlet-Multinomial Mixture framework, 2012); broader Bayesian microbiome modeling community	Faust, Raes, Friedman, Alm and colleagues
Type	Probabilistic/Bayesian pipeline for compositional count data	Integrative bioinformatics pipeline
Source fondatrice	Holmes, I., Harris, K., & Quince, C. (2012). Dirichlet Multinomial Mixtures: Generative Models for Microbial Metagenomics. PLOS ONE, 7(2), e30126.	Friedman, J., & Alm, E. J. (2012). Inferring correlation networks from genomic survey data. PLoS Computational Biology, 8(9), e1002687.
Alias	Bayesian microbiome profiling, Dirichlet-Multinomial microbiome analysis, Bayesian alpha/beta diversity, probabilistic microbiome diversity	microbial co-occurrence network analysis, microbiome network ecology, ecological network-based diversity, NBMDA
Apparentées	5	5
Résumé	Bayesian microbiome diversity analysis applies probabilistic models — chiefly Dirichlet-Multinomial and related hierarchical frameworks — to 16S rRNA or shotgun metagenomic count data to estimate alpha-diversity (within-sample richness and evenness) and beta-diversity (between-sample compositional differences) while propagating uncertainty through the entire inference chain. Unlike frequentist rarefaction-based approaches, Bayesian methods treat taxon counts as draws from a latent composition, enabling credible intervals on diversity metrics and principled comparison across groups with unequal sequencing depth.	Network-based microbiome diversity analysis integrates graph-theoretic co-occurrence network inference with classical alpha- and beta-diversity metrics to characterize the structural organization of microbial communities. Rather than treating taxa as independent entities, the method models pairwise microbial associations as edges in a network, enabling identification of keystone taxa, community modules, and ecological interaction patterns that simple diversity indices cannot detect.