Microbial sulfate reduction by Desulfovibrio is an important source of hydrogen sulfide from a large swine finishing facility.
Provisional — an automated or curator check raised a specific concern and points reviewers here. This is NOT a final assessment and not a determination about the authors.
The main results reproduced, with only marginal, non-material deviations.
- ✓Same input data as the authors
- ✓Reported values were directly comparable
- ✓Reported values are derivable from the shared data
- ✓Any deviation was negligible
- ✓The central claim held under reproduction
- 🟡A deviation arose in the data or preprocessing
- 🟡A deviation was attributed to the published material
- 🟡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
IN PROGRESS / partial. Paper is highly reproducible on paper: all data public (PRJNA691721, 15 runs; GenBank CP072608/9) and pipeline fully specified (FLASH+Sickle v1.33+Unicycler v0.4.8 for the L4 genome; OrfM+hmmsearch+diamond for metagenome dsr; USEARCH/VSEARCH/SILVA for 16S). Control-plane cross-checks already done: chromosome length EXACT (3,544,025), 2 circular contigs confirmed, Illumina read count EXACT (4,202,750), metagenome read pairs EXACT (108,273,070). Two discrepancies flagged: (1) deposited plasmid 10,867 bp vs paper 10,876 bp = likely digit-transposition typo; (2) deposited Nanopore run 34,614 reads vs paper 69,108 reads (~half). BLOCKED on heavy compute: «our HPC» account is over per-user disk quota on BOTH home and «infra» (cannot write any file), preventing read downloads + env build + assembly/dsr runs. Awaiting operator quota fix; retrying. NOT a drop (data+code fully available). Not attempted yet: from-reads assembly rerun, dsr read counting, 16S abundances, ANI.
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 70assessed: 2026-06-18 ⛓ 2e056f9f6ab1
✎ 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
- 2026-06-18
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-19no human curator yet
- 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: opusThe study aimed to characterise the microbial community responsible for H2S production and estimate the microbial sulfate reduction rate in manure slurry from a large-scale swine finishing facility, testing whether sulfate-reducing bacteria are key drivers of hydrogen sulfide formation in this low-sulfate agricultural waste system.
- ★ Sulfate-reducing Desulfovibrio are the key microbial players responsible for H2S production in swine manure slurry despite being a minor community component finding
- ★ Microbial sulfate reduction rate in manure slurry is high (7.25 nmol S cm-3 day-1), comparable to active marine sediments finding
- ★ Sulfate reduction in manure slurry is limited by sulfate concentration and availability of low-molecular-weight electron donors such as lactate mechanism
- ★ Gypsum (CaSO4) present in manure may serve as a solid-phase electron acceptor for sulfate reduction mechanism
- ★ Metagenomic dsrA/dsrB/dsrD marker analysis identifies Desulfovibrionaceae as the dominant sulfate-reducing lineage in the manure method
- ★ A novel strain Desulfovibrio desulfuricans L4 was isolated, harbouring a plasmid (pDsulf-L4) with a multidrug-resistance cassette similar to enterobacterial plasmids resource
- Low-sulfur diet, manure treatment with iron salts, and avoiding gypsum bedding are possible mitigation strategies for H2S emissions finding
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| 35SO4 2- radiotracer sulfate reduction rate measurement | swine manure slurry (Tom3) | sulfate amendment; sulfate plus lactate amendment; none (control) | sulfate reduction rate (nmol S cm-3 day-1); AVS and CRS sulfur fractions | — |
| 16S rRNA gene amplicon profiling | sediment slurry (Tom1) and manure slurry (Tom3) microcosms | sulfate amendment; sulfate plus lactate amendment; none (control) | relative abundance of microbial taxa | — |
| Shotgun metagenomic sequencing (dsrA/dsrB/dsrD marker mapping) | manure sample from manure lagoon (Tom3) | none | taxonomic assignment of dsr reads | — |
| Targeted SRB enrichment, isolation and genome sequencing | Tom1 sediment slurry; isolates Desulfovibrio L2 and L4 | 7.5 mM formate plus zero-valent iron enrichment | 16S rRNA similarity, ANI, genome/plasmid content | Widdel-Bak (WB) medium |
| X-ray diffraction (XRD) mineralogy | sediment (Tom1) and manure slurry (Tom3) | none | mineralogical composition (gypsum, sulfur, calcite, ankerite) | — |
| SEM with X-ray microanalysis (SEM-EDS) | manure slurry solids (Tom3) | none | elemental composition of solids | — |
| Ambient air H2S monitoring and pore water ion chemistry | solid-liquid separation lagoon (Tom1) and manure storage lagoon (Tom3) | none | H2S air concentration, temperature, pH, Eh, dissolved ions | OKA-T portable gas analyser with electrochemical sensor |
- – Sulfate reduction rate measured directly in manure slurry without amendment 7.25 nmol S cm-3 day-1
- ▲ Sulfate amendment enhanced the sulfate reduction rate 427 nmol S cm-3 day-1
- ▲ Lactate plus sulfate had the most pronounced effect on SRR, increasing it nearly 100-fold over control 674 nmol S cm-3 day-1 (~100x)
- ▲ Desulfovibrionaceae abundance increased with sulfate and lactate amendment in Tom3 microcosms 0.07% (control) to 0.28% (sulfate) to 1.72% (sulfate+lactate)
- – Most dsrA reads taxonomically assigned to Desulfovibrionaceae 81.4% of dsrA reads (153/188)
- – Acid volatile sulfide (AVS) was the major product of 35SO4 2- reduction 69% of total reduced sulfur
- – Strain L4 belongs to Desulfovibrio desulfuricans, a new strain (ANI 98.62% to Essex 6, above 95% species cutoff) 99.94% 16S similarity; ANI 98.62%
- – Ambient air H2S concentration ranged seasonally, exceeding the Russian residential limit of 0.008 mg m-3 0.08 to 0.69 mg m-3
- other 7.25 nmol S cm-3 day-1 (sulfate reduction rate in manure slurry (control))
- other 674 nmol S cm-3 day-1 (SRR with sulfate + lactate amendment (~100x control))
- other 427 nmol S cm-3 day-1 (SRR with sulfate amendment)
- count 81.4% (153 reads) (dsrA reads assigned to Desulfovibrionaceae out of 188 total)
- other 17.4 mg L-1 (sulfate concentration in pore water (Tom3))
- other 98.62% (ANI between strain L4 and D. desulfuricans Essex 6 genomes)
- count 69% (AVS fraction of total reduced sulfur)
- other 16 ± 2 mg/L (Tom1); 19 ± 4 mg/L (Tom3) (H2S content reached in microcosms)
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 descriptive microbiological study characterised sulfate-reducing bacteria (SRB) and sulfate reduction rates (SRR) in swine manure slurry from two lagoons using 16S rRNA gene profiling, shotgun metagenomics, and radiotracer (35SO4²⁻) assays. Results were reported primarily as descriptive statistics (means, standard deviations, relative abundances as percentages, and raw read counts) without formal inferential hypothesis testing. Microcosm experiments compared three amendment conditions (control, sulfate, sulfate plus lactate) with triplicate analysis for the Tom3 site; outcomes were characterised narratively rather than by stated statistical tests.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| Descriptive summary (mean ± SD) of H2S air concentrations | Figure 2: time-course monitoring of ambient air H2S at Tom1 | Three readings per time point | not stated |
| Descriptive summary (mean ± unspecified dispersion measure) of H2S in microcosms | H2S content in Tom1 and Tom3 microcosms (16 ± 2 mg/L and 19 ± 4 mg/L) | Not explicitly stated for this comparison | not stated |
| Informal/narrative comparison of relative abundances across microcosm conditions | Figure 5: 16S rRNA gene profiling of Tom1 and Tom3 microcosms (control vs sulfate vs sulfate+lactate) | Tom3 in triplicate; Tom1 replication not stated | not stated |
| Average nucleotide identity (ANI) threshold comparison against species cutoff | Genome-level classification of isolate L4 vs Desulfovibrio desulfuricans Essex 6 (98.62% vs 95% cutoff) | na | na |
| Proportional/count summary of metagenomic dsr reads by taxonomic lineage | Table 2: dsrA, dsrB, dsrD read taxonomic assignment | 188 dsrA, 156 dsrB, 24 dsrD reads total | not stated |
-
Microbial community composition across amendment conditions (control, sulfate, sulfate+lactate) was compared by inspecting relative abundance percentages narratively↳ Could also: Permutational multivariate ANOVA (PERMANOVA) on a Bray-Curtis or UniFrac distance matrix, or differential abundance testing (e.g., DESeq2, ANCOM-BC) with FDR correction, could also be applied to the 16S read count table — These methods would provide a formal test of whether community composition or taxon-level abundance differed significantly across amendment conditions, and would quantify effect sizes with adjusted p-values to account for the many taxa tested simultaneously
-
Sulfate reduction rates under different amendment conditions were reported as single numeric values (7.25, 427, 674 nmol S cm⁻³ day⁻¹) without stated replication or inferential test↳ Could also: Replicated radiotracer incubations with a one-way ANOVA or Kruskal-Wallis test (depending on distributional assumptions) and a post-hoc multiple comparison correction (e.g., Tukey HSD or Dunn's test) could also be used — Replicated measurements with formal testing would allow quantification of variability around each SRR estimate and a statistically supported statement about whether the amendment-driven differences are larger than within-condition noise
-
H2S concentrations in microcosms were summarised as mean ± an unspecified dispersion measure (± 2 and ± 4 mg/L)↳ Could also: Explicitly labelling the dispersion measure as SD or SEM, or reporting a 95% confidence interval, is also standard practice — SD conveys the spread of individual observations around the mean, while SEM conveys precision of the mean estimate; CI directly supports inference; making this explicit allows readers to correctly interpret the variability reported
-
Relative Sulfurimonas abundance across Tom3 conditions was described as 'not significantly different' without citing a statistical test↳ Could also: A Fisher's exact test or chi-squared test on read counts, or a negative-binomial model (e.g., DESeq2 Wald test), could also formally evaluate this claim — Formal testing would provide a p-value and effect size to support the stated conclusion, and would make the basis for the inference transparent and reproducible
-
Alpha diversity of microbial communities was not reported; communities were characterised by listing dominant phyla and their percentage ranges↳ Could also: Shannon entropy, observed ASV richness, or Faith's phylogenetic diversity indices could also be calculated and compared across samples — Diversity indices summarise community richness and evenness in a single comparable metric and would complement the percentage-based overview, particularly for comparing Tom1 and Tom3 before and after amendments
-
The correlation between ambient H2S concentration and outdoor temperature (Fig 2) was described as a tendency without a formal correlation statistic↳ Could also: A Spearman rank correlation (given the small seasonal dataset and non-normal distribution likely for both variables) or a simple linear regression could also quantify this relationship — A correlation coefficient and its confidence interval would allow readers to assess the strength and direction of the temperature–H2S association quantitatively rather than visually
What was reproduced
The exact results taken into scope, with each reported value next to the value our attempt produced.
Scope — pmid-34021225
Paper: Karnachuk et al. 2021, Sci Rep 11:10720. "Microbial sulfate reduction by Desulfovibrio is an important source of hydrogen sulfide from a large swine finishing facility." DOI 10.1038/s41598-021-90256-w · PMCID PMC8140134.
Named code pointer: https://github.com/najoshi/sickle (Sickle v1.33, a read trimmer). Per BRIEF rule P16, a third-party tool applied to the paper's own data is a fully valid reproduction. Sickle is one step inside the L4 genome assembly pipeline below.
Data: BioProject PRJNA691721 (15 SRA runs) + GenBank CP072608 / CP072609.
Datasets actually present in PRJNA691721 (from SRA runinfo)
| group | runs | platform | strategy | notes |
|---|---|---|---|---|
| 16S amplicon | SRR13442334–345 (12 runs) | MiSeq 2×300 | AMPLICON | 3 Tom1 + 9 Tom3 samples |
| metagenome | SRR13442943 (1 run) | HiSeq2500 2×150 | WGS metagenomic | Tom3 manure, 108,273,070 spots |
| L4 genome Illumina | SRR13452044 | "HiSeq2500" 2×~243 | WGS genomic | 4,202,750 spots — matches paper exactly |
| L4 genome Nanopore | SRR13452043 | MinION | WGS genomic | 34,614 reads (paper says 69,108 — see claim G5) |
IN SCOPE (pipeline-derived, attempted)
A. Desulfovibrio sp. L4 genome hybrid assembly ← PRIMARY (uses sickle)
Pipeline (Methods, "Sequencing of … L4 genome"): FLASH v1.2.11 merge → Sickle v1.33 trim → Unicycler v0.4.8 hybrid assembly of Illumina (SRR13452044) + Nanopore (SRR13452043) → RAST annotation.
- Reported outputs: 2 circular contigs — chromosome 3,544,025 bp, plasmid pDsulf-L4 10,876 bp; plasmid coverage 1885× = 6.2× the chromosome.
- Falsifiable, single-genome, all-open-tools → cleanest target.
B. Metagenome dsrABD read counts (SRR13442943)
Pipeline (Methods): OrfM v0.7.1 ORF prediction (min 96 nt) → hmmsearch v3.1b2 vs TIGRFAM dsrA/dsrB + Pfam dsrD HMMs (E<1e-3) → DIAMOND v0.9 vs UniRef100 (E<1e-3) → NCBI-taxonomist for taxonomy.
- Reported outputs (Table 2 totals): dsrA=188, dsrB=156, dsrD=24 reads mapped.
- Total counts depend only on OrfM+hmmsearch (tractable). Taxonomic breakdown (Table 2 columns) needs UniRef100 (~hundreds of GB) — attempt if feasible, else flagged out of reach.
C. 16S amplicon read counts (SRR13442334–345)
Pipeline: FLASH merge → USEARCH v11 OTU clustering (97%) → VSEARCH vs SILVA v132.
- Reported: "at least 11,000 … on average 33,316 for Tom1 and 25,760 for Tom3" reads used (post-merge/QC). Raw spot counts checkable directly; post-QC counts need FLASH+USEARCH. USEARCH v11 is proprietary (free 32-bit binary).
- Relative-abundance bar charts (Fig 5) = deeper stretch target.
D. Genome-record + ANI cross-checks
- GenBank CP072608/CP072609 lengths vs reported chromosome/plasmid (direct check).
- ANI L4 vs D. desulfuricans Essex 6 = 98.62% (fastANI/pyani).
- 16S similarity L4 vs Essex 6 = 99.94%; L2 vs D. vulgaris Hildenborough 99.78%.
OUT OF SCOPE (wet-lab / instrumental / manual — NOT attempted)
- SRR (sulfate-reduction rate) 7.25 nmol S cm⁻³ day⁻¹ — radiotracer assay.
- H₂S concentrations (16±2, 19±4 mg/L), sulfate 17.4 mg/L, Fe (ICP-MS), pH/redox.
- XRD mineralogy (gypsum), SEM/EDS imaging, RAST functional gene annotation (descriptive, not a quantitative pinnable value).
Known data-vs-paper discrepancies to flag
- Nanopore read count: paper 69,108 reads / ~348 Mbp vs SRA SRR13452043 34,614 reads / ~157 Mbp. Possible second undeposited run or basecalling difference.
- L4 Illumina platform: paper says MiSeq 2×300; SRA labels HiSeq2500. Spot count (4,202,750) matches exactly, so it is the right run — metadata label mismatch only.
- Metagenome "~2 million reads" in Results vs 108,273,070 spots in SRA / Methods — to clarify against the run.
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.
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.