What it is
MetaPhlAn3 is a microbial taxonomic profiling method and namespace used in the discoveries analysis of cross-cohort microbiome ecotype projection. 1
Canonical name: MetaPhlAn3. 1
Known aliases: MetaPhlAn 3. 1
Stable external identifier: Not specified in discoveries. 1
Key facts
A linear discriminant analysis plus Gaussian mixture model (LDA-GMM) projection used MetaPhlAn3 and Kaiju namespaces and produced plausible Kuehl cohort proportions of 27%, 42%, and 31% across the inferred ecotypes. 2
By contrast, a centered log-ratio transformation plus principal component analysis and Gaussian mixture model (CLR-PCA GMM) projected all 26 Kuehl samples to the E3 ecotype at confidence greater than 0.97 because Kuehl detected only 54% of the training species. 1
These results identify namespace compatibility and species-detection coverage as important limitations when transferring microbiome ecotype models across cohorts. 1
The ibd_phage_targeting analysis supports the use of MetaPhlAn3 for ecotype construction: 8,489 curatedMetagenomicData samples were analyzed, and a consensus K = 4 framework defined four ecotypes before projection to 23 UC Davis patients. 2 The same analysis refines the earlier transferability limitation by reporting 48.9 % per-sample LDA/GMM agreement, leave-one-substudy-out ARI with mean 0.113 and range 0.000–0.282, and 50.6 % agreement after pathway-feature refitting. 2
It also supports the prior warning about method asymmetry: LDA was judged more robust, whereas GMM assigned all 26 UC Davis samples to E3 with confidence >0.97 under sparse feature overlap, an interpretation considered artifactual. 2
The pitfalls analysis refines these transferability warnings: cross-cohort MetaPhlAn3 analyses require a synonymy layer rather than simple string normalization because taxonomic names may differ across short names, full lineage strings, and GTDB reclassifications, including Bacteroides vulgatus/Phocaeicola vulgatus, Eubacterium rectale/Agathobacter rectalis, and Ruminococcus gnavus/Mediterraneibacter gnavus. 3 In the CrohnsPhage mart, failure to reconcile names produced log₂FC approximately 28, equivalent to approximately 2.68 × 10⁸ fold, because one cohort’s abundance collapsed to a pseudocount; a targeted battery used a hand-curated 23-entry synonym map, while full-taxonomy aggregation requires an NCBI-taxid-backed, GTDB-version-aware reconciliation layer. 3
Related pages
- discoveries — source summary containing the MetaPhlAn3 projection findings.
- Metagenome-Prioritized Phage Cocktails for Crohn’s Disease and IBD — source summary of the ecotype, multi-omics, and phage-targeting analysis. 2
- pitfalls — source summary documenting MetaPhlAn3 name-reconciliation and cross-cohort analysis pitfalls. 3
- curatedmetagenomicdata — dataset used in the cross-cohort ecotype analysis. 1
- multi-omics-integration — cross-cohort and multi-modal integration context for microbiome analyses. 1