Taxonomic and functional partitioning of Chloroflexota populations under ferruginous conditions at and below the sediment-water interface.
Part of the results reproduced; minor but material deviations remained.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
- ✓Same input data as the authors
- ✓Reported values were directly comparable
- ✓The central claim held under reproduction
- 🟡A deviation arose in the data or preprocessing
- 🟡A deviation was attributed to the published material
- 🟡Reported values were not (fully) derivable from the shared data
- 🟡The deviation was non-trivial in magnitude
- 🟡Overall, the reproduction showed a material discrepancy
A 0–100 reproducibility-quality score from the per-question grades, shown as a z-score: standard deviations above (+) or below (−) the mean of comparable assessments.
▸Reproduction agent’s raw note
salvaged by watchdog from agreement.json (agent omitted ROOM_RESULT.json)
These records describe the outcome of reproduction attempts carried out autonomously by brainbox using large language models (LLMs). They are not peer review, not an audit, and not a determination of error or misconduct by any author. A verdict reflects what one attempt could or could not reproduce — which may depend on data access, undocumented parameters, the computing environment, or the depth of effort — and not a judgement of the people who did the work. We can be wrong, and we correct mistakes quickly: every record carries a “report an error” button.
✎ I am an author of this paper
Updated or fixed a deposit, or is there an erratum? Ask us to re-run the metrics. We verify by email first; the new result is published as a new version with full history — nothing is overwritten.
Provenance — full disclosure
When this reproduction was carried out, which methodology version was used, and by whom — so the record can be audited and checked independently.
- Reproduced
- 2026-06-16
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-16no human curator yet
- Last updated
- 2026-07-29
Provisional, curator- or AI-assessed, and independently checkable. A reproduction outcome states what one attempt could reproduce — not a judgement of the authors.
Deep full-text extraction
Model: opusThe authors hypothesize that oligotrophic anoxic ferruginous (iron-rich, sulfate-poor) settings such as Lake Towuti and its sediments constitute a preferential ecological niche for investigating metabolic versatility in modern Chloroflexota, testing whether Chloroflexota populations partition taxonomically and functionally according to alternative electron acceptors and donors available at and below the sediment-water interface as an analog of primeval ferruginous ecosystems.
- ★ Chloroflexota populations are partitioned according to alternative electron acceptors (Anaerolineae) and electron donors (Dehalococcoidia) among respiratory and fermentative metabolites at and below the sediment-water interface finding
- ★ Chloroflexota benefit from cross-feeding on metabolites derived from canonical respiration chains and fermentation finding
- ★ Anaerolineae have metabolic potential to use unconventional electron acceptors, different cytochromes, and multiple redox metalloproteins to cope with oxygen fluctuations, enabling colonization of the ferruginous sediment-water interface mechanism
- ★ In sediments, Dehalococcoidia evolved as acetogens scavenging fatty acids, haloacids, and aromatic acids, bypassing specific carbon assimilation steps to perform energy-conserving secondary fermentations combined with CO2 fixation via the Wood–Ljungdahl pathway mechanism
- ★ A clear taxonomic transition occurs at ~7 cmblf, with decreasing Anaerolineae and increasing Dehalococcoidia abundances with depth finding
- ★ Functional marker genes show metabolic potential for ammonia-forming nitrite reduction (nrfA) in the upper 5 cmblf and dissimilatory sulfate reduction (dsrA) below, with dmsA and arsC indicating reduction of alternative electron acceptors finding
- ★ 16 metagenome-assembled genomes (MAGs) assigned to Chloroflexota were recovered and analyzed for metabolic potential resource
- Chloroflexota would have developed analogous primeval features due to oxygen fluctuations in ancient ferruginous ecosystems mechanism
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| 16S rRNA gene amplicon sequencing (V4, 515F/806R) | Lake Towuti anoxic ferruginous sediments (upper 55 cmblf), Indonesia | none | ASVs / relative abundance and richness of Chloroflexota taxa | Illumina NovaSeq (2 × 250 bp); DADA2 v1.20; SILVA 138 |
| Shotgun metagenomics / MAG reconstruction | Lake Towuti sediment (8 samples: 0–1, 2–4, 6–8, 10–12, 14–16, 20–25, 30–35, 40–45 cmblf) | none | contigs, ORFs, functional marker genes, 16 Chloroflexota MAGs with completeness/contamination | NovaSeq 6000 (2 × 150 bp); Nextera XT kit; ATLAS v2.1.0 / metaSPAdes / GTDB v2.1.1 |
| Pore water geochemistry (ion chromatography, ferrozine spectrophotometry) | Lake Towuti sediment pore water | none | SO4 2-, NH4+, dissolved Fe2+ concentrations | non-suppressed/suppressed ion chromatography; DR 3900 spectrophotometer (Hach), absorbance 562 nm |
| Methane concentration (gas chromatography) | Lake Towuti sediment | none | CH4 concentration | Thermo Finnigan Trace gas chromatograph |
| Sulfate reduction rate measurement (35SO4 2- radiotracer incubation) | undisturbed Lake Towuti sediment mini-cores | 35SO4 2- tracer addition | sulfate reduction rates (SRR) | cold chromium distillation; Tri Carb 2500 TR liquid scintillation counter |
| Total cell counts (epifluorescence microscopy) | Lake Towuti sediment (formalin-fixed) | none | total cell counts; counts normalized to Chloroflexota 16S relative abundance | SYBR Green I staining; Leica DM2000 microscope |
| Water column profiling (CTD) | Lake Towuti water column (to 156–200 m depth) | none | pH, O2, Fe2+, SO4 2- depth profiles | Sea-Bird SBE-19 CTD probe; Niskin bottles |
| Phylogenetic / functional gene analysis (16S rRNA and RpoD; FeGenie, BLASTp) | Chloroflexota ASVs and ORFs from Lake Towuti | none | phylogenetic trees, co-occurrence network, presence/absence of functional marker genes (nrfA, dsrA, dmsA, arsC, sodM, nifU, RuBisCO, etc.) | RAxML/ARB; PhyML BLOSUM62 (SeaView); DIAMOND v0.9.24; FeGenie |
- – Taxonomic assemblage shows a clear transition at ~7 cmblf, with a decrease in Anaerolineae and an increase in Dehalococcoidia abundance with depth
- – nrfA (ammonia-forming nitrite reduction) present in the upper 5 cmblf, while dsrA (dissimilatory sulfite/sulfate reduction) present below
- – dmsA (DMSO reductase) and arsC (arsenate reductase) detected, indicating metabolic potential for reduction of alternative electron acceptors
- – sodM (superoxide dismutase) present in anaerobic Chloroflexota, suggesting defense against oxidative stress from oxygen fluctuations
- – 4 MAGs are high quality (≥90% completeness, ≤5% contamination) and 9 MAGs are good quality (≥70% completeness, ≤10% contamination) out of 16 Chloroflexota MAGs 4 high-quality, 9 good-quality of 16
- – Form I RuBisCO large subunit detected but not associated with any autotrophic pathway
- – Sulfur, iron, and methane biogeochemical processes co-occur in Lake Towuti sediments despite extremely low sulfate (<20 µM) sulfate <20 µM
- – Dehalococcoidia metabolic potential supports acetogenesis from fatty acids, haloacids, and aromatic acids combined with Wood–Ljungdahl CO2 fixation
- count 631 ASVs (16S rRNA genes) assigned to Chloroflexota (basis of PCoA and phylogenetic tree)
- count 18 915 ORFs assigned to Chloroflexota (metagenome ORFs used in PCoA)
- count 16 MAGs assigned to Chloroflexota (genomes analyzed for metabolic potential)
- count 128 ASVs retained for network construction after quality filtering (co-occurrence network from 631 ASVs)
- correlation Spearman coefficient >0.6; P value <0.01 (network analysis filtering threshold)
- other sulfate concentration <20 µM (Lake Towuti sediment/pore water)
- other detection/quantification limits 2.0 and 8.4 µM for SO4 2-; 11.3 and 67.6 µM for NH4+ (ion chromatography method limits)
- other RpoD conserved region of 167 amino acids (PhyML phylogenetic tree of RNA polymerase sigma 70 factor)
Statistical methods review
Model: sonnetA neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.
This observational metagenomics study characterizes Chloroflexota communities across a sediment depth profile in a ferruginous lake using 16S rRNA amplicon sequencing, metagenome-assembled genomes (MAGs), pore water geochemistry, cell counts, and sulfate reduction rates. Community structure was assessed by Principal Coordinates Analysis (PCoA) on both ASV and ORF datasets using the Bray–Curtis dissimilarity index, and co-occurrence network structure was resolved via Spearman correlation followed by a walktrap modularity algorithm. Phylogenetic relationships were inferred by maximum likelihood methods; results were reported primarily as relative abundances, presence/absence of functional genes, and visual ordinations, with no formal between-group hypothesis tests or reported effect sizes.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| Principal Coordinates Analysis (PCoA) with Bray–Curtis similarity index | Beta diversity of Chloroflexota 16S rRNA ASVs and metagenomics ORFs across sediment depth intervals (Figure 1d) | 631 ASVs; 18 915 ORFs | not stated |
| Spearman correlation (coefficient >0.6, P <0.01) for co-occurrence network construction | Filtering of 16S rRNA ASV dataset to 128 ASVs for network analysis | 631 ASVs pre-filter; 128 ASVs post-filter | not stated |
| Walktrap community detection algorithm (modularity) | Co-occurrence network modularity among 128 ASVs | 128 ASVs | na |
| RAxML maximum likelihood phylogenetic inference (100 replicates, maximum parsimony for best-tree selection) | 16S rRNA partial gene phylogeny of Chloroflexota ASVs (Figure 2a) | 631 ASVs | not stated |
| PhyML maximum likelihood phylogenetic inference (BLOSUM62 model, 100 bootstrap replicates) | RpoD protein phylogeny of MAG-assigned ORFs (Figure 2b) | 16 MAGs | not stated |
| BLASTp / DIAMOND similarity search (taxonomic and functional annotation) | Functional marker gene identification in contigs and MAGs (Figures 3–5) | 18 915 ORFs; 37.8 million reference proteins | na |
-
Beta diversity was visualized with PCoA (Bray–Curtis) without a formal permutational test of group separation↳ Could also: PERMANOVA (adonis2 in vegan) or ANOSIM applied to the same Bray–Curtis distance matrix could quantify whether depth zones or taxonomic classes differ significantly in community composition — Ordination plots show clustering visually but do not yield a p-value or R² effect size; a permutation test would provide a formal measure of the variance explained by the grouping factor alongside the ordination
-
Co-occurrence network edges were filtered using a fixed Spearman correlation threshold (|r| >0.6, p <0.01) without correction for compositional data bias↳ Could also: Compositionality-aware correlation methods such as SparCC, SPIEC-EASI, or proportionality (propr) could also be applied to relative abundance data — 16S rRNA relative abundances are compositional (sum-to-one constraint), which induces spurious correlations in standard Pearson/Spearman analyses; methods designed for compositional data reduce false-positive co-occurrence edges
-
Multiple Spearman correlations across 631 ASV pairs were computed for network construction with a fixed p-value threshold of 0.01 but no multiplicity correction↳ Could also: A Benjamini–Hochberg FDR correction applied to the family of pairwise correlation p-values would also control the expected proportion of false-positive edges — With hundreds of ASVs, the number of pairwise tests is large, and a per-comparison α of 0.01 may still yield many false positives; FDR correction is a standard approach in network ecology to account for this
-
Relationships between pore water geochemistry and community composition were described visually by overlaying geochemical profiles with relative abundance plots↳ Could also: Distance-based redundancy analysis (db-RDA) or a Mantel test with the geochemical variables as constraints could also be used to formally quantify how much of the community variation is explained by each geochemical gradient — db-RDA (capscale in vegan) links the ordination directly to measured environmental predictors and yields R², F-statistics, and permutation p-values, providing a quantitative alternative to visual co-inspection of profiles
-
Alpha diversity of the 16S rRNA dataset was summarized as richness (% ASVs) only↳ Could also: Shannon entropy, Simpson's index, or phylogenetic diversity (Faith's PD on the RAxML tree) would also describe evenness and phylogenetic breadth in addition to species richness — Richness counts ASVs but does not weight by their relative contribution to the community; Shannon and Faith's PD capture complementary dimensions of diversity that can differ meaningfully across depth intervals
-
MAG quality thresholds (completeness ≥70–90%, contamination ≤5–10%) were applied via CheckM, and gene presence/absence was reported as a binary bubble plot without statistical comparison between MAG classes↳ Could also: A Fisher's exact test or chi-square test on the presence/absence matrix across Anaerolineae vs. Dehalococcoidia MAGs could also be used to identify genes significantly enriched in one class; alternatively, pangenome tools (e.g. Roary, PIRATE) provide statistical frameworks for accessory vs. core gene comparisons — Binary presence/absence comparisons across two phylogenetic groups lend themselves to formal enrichment testing, which would give a p-value and odds ratio for each gene category rather than a visual impression from the bubble plot
Citation network
Where this publication sits in the reproducibility-weighted citation graph — what it is built on, and what is built on it. Citation data from OpenAlex.
No assessed neighbours yet — the network grows as more papers are assessed.
What was reproduced
The exact results taken into scope, with each reported value next to the value our attempt produced.
Scope — pmid-39384533
Paper: Vuillemin et al. 2024, Taxonomic and functional partitioning of Chloroflexota populations under ferruginous conditions at and below the sediment–water interface. FEMS Microbiol Ecol 100:fiae140. PMID 39384533 · PMCID PMC11650866 · DOI 10.1093/femsec/fiae140
- Data: ENA PRJEB66721 — 16S rRNA gene amplicons (Illumina MiSeq 2×250,
V4, 515F/806R) for 18 sediment depths (TM0-1 … TM50-55) + 2 mock communities
- 3 negative/PCR controls; and 8 shotgun metagenomes (NovaSeq 2×150, ~16–20 Gbp each, ~140 GB raw total).
- Code: https://github.com/williamorsi/MetaProt-database — custom Python/Perl scripts that turn DIAMOND BLASTp hits against the 32 GB MetaProt protein DB (SEED+MMETSP+RefSeq, hosted separately at LMU Open Data) into gene-abundance matrices with COG/KOG annotation. Authors' own functional-annotation scripts.
Pipeline-derived results & in/out-of-scope decision
| # | Reported result | Pipeline | In scope? |
|---|---|---|---|
| R1 | 631 Chloroflexota ASVs = 268 Dehalococcoidia + 328 Anaerolineae + 35 minor classes; total ASV set from 16S | Cutadapt v3.5 → DADA2 v1.20 (pooled, truncLen c(220,180)) → SILVA 138 taxonomy | YES — primary target. Small data (~40 MB), fully specified params, third-party tool (DADA2) on paper's own data (P16-valid). |
| R2 | 16 Chloroflexota MAGs (4 HQ / 9 good / 3 medium); 9 Anaerolineae, 6 Dehalococcoidia, 1 TK10 | ATLAS v2.1.0 (metaSPAdes, MetaBAT2, MaxBin2, DAS Tool, CheckM) + GTDB v2.1.1 | NO — the hard 20%. Requires assembly+binning of 8× NovaSeq (~140 GB raw); days of compute. Skipped, documented. |
| R3 | 18,915 Chloroflexota ORFs; functional gene relative abundances (‰ ORFs): nrfA, dsrA, sodM depth profiles (Fig 3b) | DIAMOND BLASTp vs 32 GB MetaProt DB → MetaProt-database scripts → CAZy/FeGenie | NO — the hard 20%. Depends on R2 assemblies + an external 32 GB protein DB; very heavy. Skipped, documented. |
| R4 | PCoA variance (37.8%/16.7% taxonomic; 41.9%/19.5% functional); network (128 nodes, 2838 edges, density 0.3492) | Downstream R on R1/R3 outputs | NO. Depends on full R1+R3 matrices. |
| — | Geochemistry, cell counts, sulfate-reduction rates (log10 cells, nmol cm⁻³ d⁻¹) | Wet-lab / PANGAEA #861437 | OUT — non-pipeline. Not attempted. |
Plan
Reproduce R1 end-to-end on «our HPC»/SLURM with «infra» data: fetch the 18
environmental 16S MiSeq runs from ENA, remove 515F/806R primers + quality-trim
with Cutadapt (-e 0.2 -q 15 -m 150 --discard-untrimmed), run DADA2 (pooled,
truncLen=c(220,180), maxN=0, rm.phix=TRUE, minLen=160), assign taxonomy with
SILVA 138, drop chloroplasts/mitochondria/singletons, then count total ASVs and
the Chloroflexota class breakdown. Compare to 631/268/328/35.
R2–R4 are explicitly not attempted (compute cost + external 32 GB DB); this is a partial reproduction by design (80/20 rule).
Assessments & scoring basis
Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.
An automated assessment. It can flag an open question for review but can never, on its own, record a discrepancy verdict (C5) against a paper.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
Same public ENA data (PRJEB66721) and the paper's stated DADA2/SILVA pipeline were run end-to-end, so data identity and endpoint comparability are clean (green). Absolute Chloroflexota ASV counts came out a uniform ~25% high (631→788, 268→337, 328→419) while the minor class matched (35→32) and the Dehalo/Anaero ratio is essentially identical (0.804 vs 0.817), so the central partitioning claim fully holds. The offset sits in upstream preprocessing — most likely the unreported maxEE filter (we used Inf) plus a SILVA 138→138.1 version difference — which is a shared methodology/underspecification issue, not fabrication or an authors' data defect. Verdict: solid partial reproduction (yellow); values are derivable in kind but not 1:1, pending the maxEE=c(2,2) sensitivity run.
Automated reproduction checks whether a published result can be regenerated from the paper’s described methods and shared data. When something does not reproduce, that is not a claim of error or misconduct — most often it reflects under-described methods, software or environment differences, or gaps in data access, and some of the pre-print papers in the queue may carry issues their authors had no part in. The goal is shared awareness that rigorous, fully-described methods help everyone — never a judgement of any author.
Are you an author? We would genuinely like to hear from you — to clarify the record, add data or code, re-run the pipeline after an accession update, and publish your response right next to the assessment. Everything here is open and auditable.
🚩 Report an error in this record
Spotted something wrong — a verdict you’d contest, a data or value error, or a private detail that slipped through? Tell us, with a short justification. Authors and readers are equally welcome to write in; we review every report.
Prefer email, or the form below not working? Contact us at support@doesitreproduce.com.
Reproduction footprint
claude-opus-4-8Measured resources invested to assess this paper — sanitised (machine class only, no job ids/paths). Compute = HPC accounting (SLURM); tokens = the AI agent's session.