henhaobio
Sign in

Nucleic Acids ResearchMicrobiologyPeer-reviewed journal article

Metalog: curated and harmonised contextual data for global metagenomics samples

Michael KuhniD, Thomas SchmidtiD, Pamela FerrettiiD, Anna M Glazek, Shahriyar Mahdi RobbaniiD, Wasiu A. AkanniiD, Anthony FullamiD, Christian SchudomaiD, Ela Cetin, Mariam HassaniD, Kasimir Noack, Anna M. SchwarziD, Roman ThielemanniD, Leonie Thomas, Moritz von Stetten, Renato AlvesiD, Anandhi IyappaniD, Ece KartaliD, Ivan KeliD, Marisa Isabell KelleriD, Oleksandr M. MaistrenkoiD, Anna MankowskiiD, Suguru NishijimaiD, Daniel PodlesnyiD, Jonas SchilleriD, , Peer Bork BiD

PDFDOIPreprintData
In-depth full-text digest available · Jump to digest ↓

Abstract

Metagenomic sequencing enables the in-depth study of microbes and their functions in humans, animals and the environment. While sequencing data is deposited in public databases, the associated contextual data is often not complete and needs to be retrieved from primary publications. This lack of access to sample-level metadata like clinical data or in situ observations impedes cross-study comparisons and meta-analyses. We therefore created the Metalog database, a repository of manually curated metadata for metagenomics samples across the globe. It contains 73,082 samples from humans (including 58,506 of the gut microbiome), 10,703 animal samples, 5,146 ocean water samples, and 21,802 samples from other environmental habitats such as soil, sediment, or fresh water. Samples have been consistently annotated for a set of habitat-specific core features, such as demographics, disease status and medication for humans, host species and captivity status for animals, and filter sizes and salinity for marine samples. Additionally, all original metadata is provided in tabular form, simplifying focused studies e.g. into nutrient concentrations. Pre-computed taxonomic profiles facilitate rapid data exploration, while links to the SPIRE database enable genome-based analyses. The database is freely available for browsing and download at https://metalog.embl.de/.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (14 Aug 2025) · CC BY 4.0

Metalog manually curates and harmonises contextual metadata for 110,733 metagenomic samples from humans, animals, ocean water and other environments, linked to taxonomic profiles and SPIRE genomes.

Sample-level context such as clinical status, medication or in situ measurements is frequently incomplete in sequence archives, scattered across papers and labelled inconsistently, which hampers cross-study meta-analyses. Existing metadata databases are often narrow in scope or stale, and none manually annotates both host-associated and environmental metagenomes, a gap Metalog fills.

Key results

  1. ShownMetalog holds manually curated metadata for 110,733 metagenomic samples drawn from 838 studies. The samples span human, animal, marine and other environmental habitats worldwide.

    73,082 human samples (58,506 gut microbiome), 10,703 animal, 5,146 ocean water and 21,802 other environmental samples (e.g., soil, sediment, fresh water); geographic distribution skews toward Western and East Asian countries. · Abstract; Introduction; Database content; Discussion

  2. ShownSeveral core fields are widely annotated within their templates, but fewer samples carry combinations of fields, which constrains analyses that need several covariates together.

    Age category annotated for 91% of human samples, captivity status for 99% of animal samples and water depth for 100% of ocean water samples; only 17,324 samples carry both exact age and BMI. · Database content; Figure legends

  3. SuggestedManual curation fixes submission errors and harmonises inconsistent labels. Each sample is split or merged to match one biological sample, drugs are mapped to ChEMBL and ATC, habitats to ENVO and UBERON, and subjects get longitudinal time points.

    Issues handled include amplicon runs mislabelled as shotgun, swapped latitude/longitude, Excel drag-down value errors and 9999 placeholders; unclear subject IDs resolved by Mash-distance clustering; first sample per subject set to time point 0. Error frequencies are not quantified. · Introduction; Annotation of samples; In-depth annotation of human samples; In-depth annotation of other sample types

  4. SuggestedA worked example links medication annotations to taxonomic profiles of adult fecal samples. It recovers drug-bacteria associations in which antibiotics are prominent.

    Linear models of log-transformed bacterial abundance on medication, adjusted for study and disease status, gave 16 drugs and drug classes with significant associations at FDR 0.01; fluoroquinolones (ATC J01MA) had the most associations, vancomycin the most positive ones. · Usage example; Figure legends

  5. ShownPrecomputed microbial data and cross-links let users apply the metadata directly to abundance-level and genome-level analyses. A website supports browsing, filtering and download.

    Taxonomic profiles from mOTUs version 3.0, SPIRE-mOTUs and MetaPhlAn 4; predicted enterotypes and fecal microbial loads for adult fecal samples; SPIRE identifiers give access to MAGs; weekly public sync with dated downloads; Open Database License. · Generation of associated microbial data; Database content; Data availability

Methods, data and reproducibility

Studies were drawn from the literature and SPIRE; metadata from BioSamples records, supplementary tables, data repositories and paper text were manually matched to sequenced samples, error-checked and mapped onto four habitat templates (human, animal, ocean water, other environmental) using controlled vocabularies such as ENVO, UBERON, ChEMBL and ATC.

  • Human-associated metagenomic samples, including gut microbiome, annotated with demographics, disease status, interventions and medication · human · n = 73,082 samples (58,506 gut)
  • Animal-associated metagenomic samples annotated with host species (NCBI Taxonomy identifiers) and captivity status · multiple animal host species · n = 10,703 samples
  • Ocean water metagenomes annotated with filter sizes, salinity, water depth and other physicochemical parameters · n = 5,146 samples
  • Other environmental metagenomes such as soil, sediment and fresh water · n = 21,802 samples
  • Curated studies contributing samples across all habitats, including consortia such as HMP and TARA · multiple species · n = 838 studies
  • Usage-example subset: adult fecal samples with disease status and medication records · human
Methods & tools
mOTUs v3.0, SPIRE-mOTUs, MetaPhlAn 4, SPIRE database, ENVironment Ontology (ENVO), UBER-anatomy ONtology (UBERON), ChEMBL, Anatomical Therapeutic Chemical (ATC) classification, NCBI Taxonomy, Mash, INSDC repositories (ENA, SRA, DDBJ), EMBL-EBI BioSamples / NCBI BioSample
Data & accessions
https://metalog.embl.de/
Reagents & environment
Taxonomic profiles computed with mOTUs version 3.0, SPIRE-mOTUs and MetaPhlAn 4; public site synchronised weekly, with dated download filenames and per-sample update timestamps.
Reproducibility gaps
No code is linked for the curation workflow or the medication usage example; sequence processing is only cited as previously described. The database changes weekly, so analyses should record the dated download file used.
Limitations (6)
  • Coverage is incomplete by necessity: not every published metagenome is included, and amplicon and metatranscriptomic datasets are excluded by design. · stated by the authors
  • Sample availability is geographically skewed toward Western and East Asian countries. · stated by the authors
  • Many clinical studies report disease status but not medication, so the authors suggest that drug meta-analyses be limited to healthy cohorts plus studies reporting some medication, where unreported drugs can be presumed absent. · stated by the authors
  • Human sample locations are often country-level and placed at population-weighted country centres, so distances between samples carry uncertainty; users should check the location_resolution field. · stated by the authors
  • The medication example adjusts only for study and disease status, ignoring demographics and co-medication; fully confounded drug-disease pairs (e.g., helminthiasis with albendazole) were dropped. · stated by the authors
  • The accuracy of the manual curation (error rates, agreement between curators, how faithful the ontology mapping is) is not measured, so users cannot estimate how much annotation error remains. · noted in this summary

Read this if you run cross-study metagenomic meta-analyses and need harmonised clinical, medication, host or environmental metadata linked to INSDC samples, taxonomic profiles and SPIRE genomes.

Citations
8
FWCI
2.39
Field percentile
top 10%

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISJiang '22Corrêa '19Kok '26Pasolli '17Oderman '26Yates '21
68% semantic matchBioinformatics
Yiqi Jiang, Yanfei Wang, Lijia Che+2 more·bioRxiv·Sep 26, 2022· 6 citations