Mucospheres produced by a mixotrophic protist impact ocean carbon cycling.
The main results reproduced: recomputed values matched the published ones within tolerance.
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
- ✓No relevant deviation in data/preprocessing
- ✓No authors-side cause for any deviation
- ✓Reported values are derivable from the shared data
- ✓Any deviation was negligible
- ✓The central claim held under reproduction
- ✓Overall, the reproduction was clean
- Every checked point held up.
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
Described well enough to reproduce the pipeline-derived part. The 16S DADA2 amplicon pipeline (repo ARBramucci/Larsson2020 @ d6da9ca, params from the script + Methods) was run end-to-end on the paper's own SRA data PRJNA737517 (21 runs) on «our HPC». Result is largely 1:1 with tolerance: the reported rarefaction depth of 23,960 reads (95% of the lowest per-sample bacterial total, after dropping an under-sequenced replicate) reproduces to 23,013 (~4% lower), and the reported value is bracketed by the repro per-sample 95%-of-min range (23,225-24,132). The under-sequenced-replicate removal step also reproduces, though our run (dada2 1.38 vs the paper's 1.14) flags 2 such samples vs 1. ASV count (314) and organellar/non-Bacteria filtering reproduce qualitatively (paper reports no ASV count). NOT attempted: the NMDS/PERMANOVA xenic-vs-axenic comparison (S5), because the per-SRR xenic/axenic+ group labels are absent from the deposited SRA metadata - a metadata reproducibility gap, not fabrication. OUT OF SCOPE (not pipeline-derived, not attempted): all carbon-export, mucosphere-production, sinking-velocity, microscopy and flow-cytometry results. No fabrication detected: the headline 16S number is genuinely derivable from the shipped data. Deviation: could not build a runnable dada2 1.14.0 conda stack (ShortRead/S4Vectors S4 mismatch), used coherent current Bioconductor 3.21.
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.
Assessment versions
Every reproduction run is kept as an immutable version — anchored to the data as it stood, with a tamper-evident chain hash. A rerun (e.g. after an author updates a deposit) adds a new version; the previous one stays on record.
-
v1 current initial assessment Score 58assessed: 2026-06-15 ⛓ f34c66c4bf62
✎ 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
- 2026-06-15
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-15👤 1 human curator(s) · Level L2 2026-06-15
- Last updated
- 2026-08-05
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: opusDoes the sophisticated foraging behaviour of the widespread mixotrophic dinoflagellate Prorocentrum cf. balticum—specifically its production of carbon-rich 'mucospheres'—represent an overlooked yet quantitatively significant mechanism influencing oceanic carbon fluxes and the biological pump?
- ★ P. cf. balticum constructs carbon-rich mucospheres that attract, capture and immobilise microbial prey to facilitate phago-heterotrophic consumption finding
- ★ Abandoned, negatively buoyant mucospheres contribute an estimated 0.02–0.15 Gt of POC annually to the biological pump (0.1–0.7% of total euphotic-zone export) finding
- ★ Mucospheres actively chemoattract both prokaryotic and eukaryotic prey rather than relying on passive interception mechanism
- ★ P. cf. balticum is an obligate phototroph, facultative phago-heterotroph and constitutive mixotroph that cannot survive purely heterotrophically finding
- ★ Mucosphere production is triggered by prey-derived exometabolites and modulated by prey type, phosphorus status and light availability mechanism
- P. cf. balticum is a cosmopolitan marine eukaryote present at 96% of Tara Oceans surface stations sampled finding
- Mucosphere composition is analogous to transparent exopolymeric particles (TEP), staining for acidic polysaccharides and protein finding
- Coupling culture measurements of mucosphere carbon content and production with field abundance provides a method to estimate protist-driven carbon export method
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| 18S rDNA amplicon biogeography analysis | Tara Oceans global surface 5–20 µm size fraction samples | none | relative abundance and presence of P. cf. balticum 18S sequence across stations | Tara Oceans amplicon dataset |
| ITS/phylogenetic sequencing | four clonal P. cf. balticum strains from Port Hacking, Australia | none | Maximum Likelihood/Bayesian phylogenetic placement | — |
| Toxin screening (LC-MS/MS) | xenic P. cf. balticum cultures with associated microbiome | none | presence of okadaic acid, dinophysistoxin-1/-2, tetrodotoxin | liquid chromatography-tandem mass spectrometry (LC-MS/MS) |
| Fluorescence/confocal microscopy feeding assays | P. cf. balticum co-cultured with Rhodomonas salina and other protistan prey | prey addition (co-culture) | peduncular feeding, intracellular green-fluorescent food vacuoles, myzocytosis | Nikon Eclipse Ti inverted fluorescence microscope; Nikon AR1 confocal microscope |
| Fluorescently-labelled bacteria ingestion assay | P. cf. balticum cultures | co-culture with labelled prokaryotes (6 h) | intracellular labelled bacteria confirming prokaryote consumption | — |
| Growth experiment under trophic/light conditions | P. cf. balticum co-cultures ± prokaryotic and/or eukaryotic prey (Proteomonas sulcata) | light vs dark; ± prokaryotic/eukaryotic prey | cell density / growth over time (n=6) | — |
| Chemotaxis assay (Chemotactic Index) | prokaryotic microbiome and R. salina prey responding to mucosphere chemical extracts | mucosphere-derived chemical extract vs culture filtrate control | chemotactic index (Ic), fold enrichment of attracted cells | — |
| Mucosphere production monitoring + organic carbon quantification (TEP staining) | axenic single P. cf. balticum cells (n=26) under varied P/light/prey | P replete vs deplete (organic vs inorganic phosphate), light level, prey type | % cells producing mucospheres, number/timing of mucospheres, carbon content per mucosphere | Alcian Blue and Coomassie Brilliant Blue staining |
- ▲ Discarded mucospheres contribute an estimated 0.17–1.24 mg m−2 d−1 of POC, equating to 0.02–0.15 Gt annually to the biological pump 0.17–1.24 mg m−2 d−1; 0.02–0.15 Gt yr−1
- ▲ Estimated mucosphere carbon export of 0.04–0.29 mg C m−2 d−1, reaching as high as 7.13 mg C m−2 d−1 0.04–0.29 (up to 7.13) mg C m−2 d−1
- ▲ Mucosphere-derived chemicals attracted significantly more prokaryotic and eukaryotic prey than controls 3-fold (prokaryotes), 2-fold (eukaryotes)
- – P. cf. balticum did not grow under dark/heterotrophic conditions with prey, confirming obligate phototrophy
- ▲ Prey presence drove mucosphere production: 12% (prokaryotic), 46% (eukaryotic), 69% (both) of P-deplete cells; 42% under P-replete with eukaryotic prey 12% / 46% / 69% / 42%
- ▼ Low light suppressed mucosphere production (23% of cells vs 46% in optimal light with eukaryotic prey) 23% vs 46%
- – P. cf. balticum occurred at 96% of Tara Oceans stations (95–99) in surface 5–20 µm samples 96% of stations
- – Each cell constructed ~1 mucosphere per day and consumed a single eukaryotic prey before abandoning it ~1 mucosphere/day
- other 154.3 (± 19.5) pg carbon per mucosphere (mean carbon content per P. cf. balticum mucosphere)
- pvalue p = 9.27e−04 (two-sided t-test, prokaryotic prey chemotaxis to mucosphere chemicals)
- pvalue p = 4.07e−05 (two-sided t-test, eukaryotic prey (R. salina) chemotaxis to mucosphere chemicals)
- pvalue p = 4.06e−12 (Kruskal–Wallis test for cumulative mucosphere production across resource conditions)
- count 19 and 137 cells L−1 (max 3350 cells L−1) (lowest/highest average and maximum field abundance of P. cf. balticum across Australian time-series stations)
- other 23% daily mucosphere production rate (production under low light (20 µmol m−2 s−1) with eukaryotic prey, used for export estimate)
- other 12–83% of total POC flux; 0.60–1257 mg C m−2 d−1 (reference comparison: Appendicularian houses' contribution to surface POC flux)
- count 11 additional protistan taxa (3–25 µm) consumed (breadth of prey readily consumed by P. cf. balticum)
Statistical methods review
Model: opusA neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.
The study combines laboratory culture experiments, microscopy, chemotaxis assays, and field/amplicon distribution data to characterise a mixotrophic dinoflagellate and estimate its carbon-export contribution. Group comparisons were made with a two-sided t-test for chemotactic prey attraction and a Kruskal–Wallis test for mucosphere production across resource conditions, with exact p-values reported. Quantitative results were summarised as means with standard deviation or standard error and stated replicate counts, and the carbon-flux estimate was derived by combining culture-based measurements with field abundance ranges.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| two-sided (two-tailed) t-test | Chemotactic Index of prokaryotic and eukaryotic prey to mucosphere extracts vs culture filtrate controls (Fig. 3e) | n = 5 biologically independent samples | not stated |
| Kruskal–Wallis test | Cumulative mucosphere production under different resource availability conditions (Fig. 3f) | n = 26 cells | not stated |
-
Prey attraction was compared with two separate two-sided t-tests for prokaryotic and eukaryotic prey.↳ Could also: A single model (e.g., two-way analysis of variance with prey type as a factor) with a post-hoc multiplicity adjustment such as Benjamini–Hochberg or Bonferroni could also be applied. — A combined model with correction would also control the family-wise (or false-discovery) error rate across the related comparisons and partition variance attributable to prey type.
-
The t-test was applied with n = 5 biologically independent samples per comparison.↳ Could also: A non-parametric test such as the Mann–Whitney U test could also be used, as was done elsewhere in the paper with the Kruskal–Wallis test. — A rank-based test would also be appropriate at small n where the normality assumption is harder to verify, and would keep the analytic approach consistent across panels.
-
Dispersion was reported as standard error of the mean for the chemotaxis and mucosphere-production data.↳ Could also: Standard deviation or a 95% confidence interval could also be reported. — SD conveys the spread of the underlying observations and a 95% CI conveys precision of the estimate directly; both are often preferred for small n where SEM can appear to understate variability.
-
Group differences were summarised with p-values and means with dispersion.↳ Could also: Reporting effect sizes (e.g., Cohen's d, Hedges' g, or the fold-enrichment with a confidence interval) alongside the p-values could also be included. — Effect sizes with intervals would also convey the magnitude and precision of differences independent of sample size.
-
The carbon-export estimate was built from point measurements combined with low/high abundance ranges to produce a value range.↳ Could also: A formal uncertainty propagation or Monte Carlo simulation across the input distributions could also be used. — Propagating the variability of each input would also yield a probabilistic confidence range for the export estimate rather than a deterministic range.
-
Mucosphere production was monitored on 26 cells from one clonal strain.↳ Could also: A mixed-effects model treating strain/cell as a random effect, drawing on the four available clonal strains, could also be used. — This would also let the analysis account for among-strain and among-cell variation and support generalisation beyond a single clone.
Result convergence & founder nodes
Findings this paper shares with others that ran a comparable experiment. A node’s strength is how many independent papers report it (replication breadth) — not how often it is cited, so a heavily-replicated but under-cited founder still stands out.
-
Individual P. cf. balticum cells produce approximately one mucosphere per day, consuming a single eukaryotic prey before discarding the structureimaging prorocentrum-cf-balticum-culture 2022×1papers★ This paper is the founder (earliest)
-
P. cf. balticum mucospheres contribute 0.17–1.24 mg C m−2 d−1 POC to the biological pump, equivalent to 0.02–0.15 Gt C yr−1 globallyother global ocean surface up 2022×1papers★ This paper is the founder (earliest)
-
P. cf. balticum detected at 96% of Tara Oceans global surface stations in the 5–20 µm size fraction, indicating cosmopolitan distributionother ocean surface 2022×1papers★ This paper is the founder (earliest)
-
Mucosphere-derived chemicals attract 3-fold more prokaryotic and 2-fold more eukaryotic prey compared to culture filtrate controlsother prorocentrum-cf-balticum-culture up 2022×1papers★ This paper is the founder (earliest)
-
P. cf. balticum shows no growth under dark heterotrophic conditions with prey, confirming obligate phototrophyother prorocentrum-cf-balticum-culture none 2022×1papers★ This paper is the founder (earliest)
-
Low light suppresses mucosphere production in P. cf. balticum (23% of cells vs 46% under optimal light with eukaryotic prey)other prorocentrum-cf-balticum-culture down 2022×1papers★ This paper is the founder (earliest)
-
Prey presence increases mucosphere production in phosphate-deplete P. cf. balticum (12% prokaryotic prey, 46% eukaryotic prey, 69% both prey types)other prorocentrum-cf-balticum-culture up 2022×1papers★ This paper is the founder (earliest)
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.
Data lineage
The datasets this paper uses (text-mined from the full text via Europe PMC), and which other assessed papers stand on the same data. A shared dataset is a factual link — not a judgement.
What was reproduced
The exact results taken into scope, with each reported value next to the value our attempt produced.
Scope — pmid-35288549
Paper: Larsson ME et al. (2022) Mucospheres produced by a mixotrophic protist impact ocean carbon cycling. Nat Commun 13:1301. PMID 35288549 · PMCID PMC8921327 · DOI 10.1038/s41467-022-28867-8
Code: https://github.com/ARBramucci/Larsson2020 (commit to be pinned at run)
— a DADA2 amplicon pipeline (single R script Larsson2020-dada2pipline + readme.RMD).
Data: SRA BioProject PRJNA737517 — 21 paired-end Illumina MiSeq AMPLICON runs
(SRR14811642–SRR14811662), 16S rRNA V1–V3.
What this paper is about
The headline science is wet-lab / microscopy / biogeochemistry: a mixotrophic dinoflagellate (Prorocentrum cf. balticum) produces carbon-rich "mucospheres" that trap prey and sink, contributing to ocean carbon export. Those results (carbon content, sinking velocities, mucosphere production rates, grazing) are measured by microscopy, flow cytometry, elemental analysis and modelling — not by a bioinformatic pipeline.
The 16S rRNA amplicon sequencing is ancillary: it was used only to show that the associated bacterial consortium did not differ between xenic and "axenic+" algal cultures, so the algal physiology could not be ascribed to a changed microbiome.
IN SCOPE (pipeline-derived → we attempt to reproduce)
The DADA2 pipeline in the repo, applied to PRJNA737517, regenerates:
| ref | reproducible output |
|---|---|
| S1 | Per-sample read tracking (input → filtered → denoised → merged → non-chimeric) |
| S2 | The ASV table (count of ASVs after chimera removal) |
| S3 | Taxonomy via SILVA v138 at 50% cutoff; removal of non-Bacteria / chloroplast / mitochondria ASVs |
| S4 | Rarefaction depth = 23,960 reads = 95% of the lowest per-sample total, with one under-sequenced replicate removed (Methods). ← primary numeric, directly checkable |
| S5 | Bacterial community composition / NMDS + PERMANOVA: xenic vs axenic+ overlap, no significant difference (Suppl. Fig. 13b) — qualitative reproduction |
Pipeline parameters (from repo + Methods):
- R 3.6.3; dada2 1.14.0; phyloseq 1.30.0 (we pin nearest reproducible env on «our HPC»)
- Primers V1–V3: 27F
AGAGTTTGATCMTGGCTCAG/ 519RGWATTACCGCGGCKGCTG(removed with cutadapt) filterAndTrim: truncLen=c(255,250), maxEE=c(2,6), truncQ=6, minLen=50, maxN=0, rm.phix=TRUE- learnErrors nbases=1e8; mergePairs minOverlap=10, maxMismatch=1; removeBimeraDenovo (consensus, minFoldParentOverAbundance ~ default)
- Taxonomy: SILVA 138, 50% bootstrap/probability cutoff
OUT OF SCOPE (not pipeline-derived → not attempted)
- All carbon/biogeochemistry numbers (Table 2), mucosphere production (Table 3), sinking velocities, grazing, elemental (CHN) analysis — wet-lab/microscopy/modelling.
- Flow-cytometry verification of axenic status (Suppl. Figs 12, 14a) — instrument data.
- Light/fluorescence microscopy figures.
80/20 plan
- Stage the 21 SRA runs + repo on «infra» (inside «our HPC» job — compute nodes have net).
- Run the DADA2 pipeline with the documented parameters → read-tracking table, ASV count, per-sample bacterial-read totals.
- Primary check (S4): does 95% × (minimum per-sample total, after dropping the one under-sequenced replicate) ≈ 23,960? Identify which run is the dropped replicate (candidate: SRR14811649, 18,814 raw reads — the lowest).
- Secondary (S2/S3/S5): ASV count, taxonomy filtering, NMDS/PERMANOVA xenic-vs-axenic. These are softer (sample→treatment mapping not published per-SRR; best-effort).
Known reproducibility risks
- Sample metadata gap: SRA run titles are generic ("dinoflagellate metagenome"); the xenic/axenic+ assignment per SRR is not published, limiting S5 to a qualitative community-level check unless we can infer groupings.
- DADA2/SILVA exact versions may differ slightly on «our HPC» conda; denoising is deterministic given seed+version but ASV counts can shift with version → grade with tolerance.
- "23,960 = 95% of lowest" is comput
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.
The reproducible 16S DADA2 portion reproduces well: the reported rarefaction depth of 23,960 reads lands at 23,013 (~4% lower, bracketed by 23,225–24,132) on the paper's own SRA data, with no fabrication — the headline number is genuinely data-derived. The only deviations (4% depth, 2 vs 1 dropped under-sequenced replicate) are explained by a DADA2 version mismatch (1.38 vs 1.14), i.e. our-side/technical, not an authors' defect. However, the actual 16S analytical conclusion (S5 PERMANOVA xenic≈axenic+, Suppl. Fig.13b) is not testable because per-run treatment labels were never deposited, and the paper's central carbon-cycling claims are wet-lab and out of scope. Net: a solid but partial, version-explainable reproduction → yellow.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
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.