Transcriptome-Proteome Profiling in Burkholderia thailandensis during the Transition from Exponential to Stationary Phase.
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
- ✓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
Described well enough to reconstruct and reproduce. The repo is a single 90-line pandas/scipy script that Pearson-correlates RNA-Seq vs TMT-proteomics per-gene log2FC (StandardScaler step is affine, so Pearson-invariant -- confirmed). The integration inputs are not shipped; reproduced by running that logic on the authors' own deposited DE tables (RNA: GEO GSE279483 DESeq2 table, 5714 genes; protein: SI Table S3, 3033 proteins), joined on the shared BTH 'Gene ID'. RESULTS: (C1) HEADLINE Pearson r=0.4 REPRODUCES at the paper's 1-decimal precision (r=0.389 GeneID-join, n=2931; 0.381 UniProt-join, n=2677); the reported p=8.63e-102 is internally consistent with a merged n2634, which our two join keys bracket -- the order-of-magnitude p difference is an n effect, not an effect-size disagreement (paper does not report n). (C2) 928 DEGs (564/364) REPRODUCES EXACTLY from the full gene table at |log2FC|>=2 & FDR<0.05 (cutoff column is FDR, not PAdj). (C3) 832 DEPs: PARTIAL + FLAGGED -- the shipped 832-DEP list is internally consistent (all have |log2|>=0.5 & raw P<0.05) but the literal stated rule applied to the full 3033-protein table selects 937 (629/308); 105 proteins meet the published threshold yet were excluded from the DEP list by an UNSTATED additional criterion (not the FDR-confidence flag -- all 937 are High). Flagged for human review as possible under-specification, NOT asserted as fabrication; direction up>down preserved. NOT ATTEMPTED (hard-20% / out-of-scope): RNA-Seq from raw FASTQ (HISAT2->featureCounts->DESeq2) and TMT quant from raw MS (PXD056625 .raw/.mzid) -- both shipped only as derived tables; KEGG enrichment + STRING/Cytoscape PPI (external-DB/GUI, non-pipeline). All compute on «our HPC» SLURM (kubisch_std); env python3.11/pandas3.0.3/scipy1.17.1/sklearn1.9.0; repo @0c88161; large inputs kept on «infra» with SHA256 recorded in agreement.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.
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 80assessed: 2026-06-14 ⛓ 706858531d80
✎ 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
- 2026-06-14
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-15no 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: sonnetThe paper investigates the molecular mechanisms underlying the transition of Burkholderia thailandensis from exponential to stationary phase by integrating transcriptomic and proteomic profiling, testing whether transcriptome and proteome changes are correlated and whether RpoS accumulation (the canonical stationary-phase regulator) drives this transition.
- ★ 928 differentially accumulating mRNAs (564 up, 364 down) were identified between exponential and stationary phase finding
- ★ 832 differentially accumulating proteins were identified between exponential and stationary phase finding
- ★ mRNAs for benzoate degradation and O-antigen nucleotide sugar biosynthesis were elevated in stationary phase, while translation and flagellar biosynthesis transcripts were downregulated finding
- ★ Proteins involved in fatty acid degradation, butanoate metabolism, and secondary metabolite synthesis accumulated in stationary phase, while ribosomal proteins and iron-sulfur biogenesis proteins were downregulated finding
- ★ Only a modest correlation was observed between transcriptome and proteome changes finding
- ★ RpoS sigma factor was not significantly increased during stationary phase, despite being typically critical for stress-response gene expression finding
- A custom protein-protein interaction reference dataset was generated via STRING since B. thailandensis is not natively available in STRING method
- ★ Data point to distinct adaptive mechanisms in B. thailandensis, possibly including post-translational regulation, rather than canonical RpoS-driven regulation mechanism
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| bulk RNA-seq (Illumina TruSeq Stranded mRNA) | Burkholderia thailandensis E264 | growth phase transition (exponential vs stationary) | differential mRNA abundance (DEGs) | Illumina NextSeq 500 |
| TMT-based quantitative mass spectrometry proteomics | Burkholderia thailandensis E264 | growth phase transition (exponential vs stationary) | differential protein abundance (DEPs) | Q-Exactive Orbitrap MS with Ultimate 3000 RSLC LC |
| RT-qPCR | Burkholderia thailandensis | growth phase timecourse (OD600 0.1 to 3.0) | RpoS mRNA expression (relative to hgprt reference gene) | Luna one-step universal master mix |
| protein-protein interaction network analysis | Burkholderia thailandensis (in silico, upregulated proteins) | none | network centrality metrics, community/module detection | STRING database, NetworkX (Louvain method) |
| Gene Set Enrichment Analysis / KEGG pathway analysis | Burkholderia thailandensis DEGs/DEPs | none | normalized enrichment score, pathway-level regulation | fgsea (R), Pathview |
| Gene Ontology enrichment analysis | Burkholderia thailandensis DEGs/DEPs | none | overrepresented GO terms (BP, MF, CC) | custom Python/R scripts, ggplot2 |
| BlastP homology comparison of RpoS regulons | B. pseudomallei, E. coli, B. thailandensis proteomes | none | homologous RpoS-regulated proteins across species | — |
- – 928 DEGs identified between exponential and stationary phase (log2FC ≥|2|, FDR<0.05) 564 up / 364 down
- – 832 differentially accumulating proteins identified (log2FC ≥|0.5|, p<0.05)
- – Benzoate degradation and O-antigen nucleotide sugar biosynthesis mRNAs elevated in stationary phase; translation and flagellar biosynthesis mRNAs downregulated
- – Fatty acid degradation, butanoate metabolism, and secondary metabolite synthesis proteins accumulated; ribosomal and Fe-S biogenesis proteins markedly downregulated
- – Only modest correlation seen between transcriptome and proteome log2FC changes
- – RpoS was not significantly increased in stationary phase despite typically being abundant then
- count 928 DEGs (564 up, 364 down) (log2FC ≥2 or ≤-2, FDR<0.05, transcriptome)
- count 832 differentially expressed proteins (log2FC ≥0.5 or ≤-0.5, p<0.05, proteome)
- count 5,562 entries (UniProt B. thailandensis E264 protein database used for peptide matching)
- count 58 unique differentially expressed proteins (reported RpoS regulon in B. pseudomallei (2D gel/MS))
- count 35 proteins (reported RpoS regulon in E. coli)
- mean OD600 0.6 ± 0.05 (exponential), OD600 2.6 ± 0.05 (stationary) (growth phase sampling points, three biological replicates)
- other log2FC cutoff ≥2/≤-2, FDR<0.05 (significance threshold for DESeq2 transcriptome analysis)
- other log2FC cutoff ≥0.5/≤-0.5, P<0.05 (significance threshold for TMT proteomics analysis)
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 study compared exponential-phase and stationary-phase B. thailandensis E264 using three biological replicates per condition in a parallel RNA-seq and TMT-based quantitative proteomics design. Differential expression was identified by DESeq2 (RNA-seq) with Benjamini-Hochberg FDR correction and by a nominal P-value threshold within Proteome Discoverer (proteomics); results were reported as log2-fold changes. Pathway-level changes were explored by GSEA (fgsea) and GO enrichment, and transcriptome–proteome correspondence was assessed by Pearson correlation of normalized log2-fold changes.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| DESeq2 Wald test | RNA-seq differential gene expression, exponential vs. stationary phase | 3 biological replicates per condition (6 total libraries) | not stated |
| Nominal P-value threshold (<0.05) within Proteome Discoverer (specific test not named in text) | TMT proteomics differential protein abundance, exponential vs. stationary phase | 3 biological replicates per condition (6 TMT channels) | not stated |
| Pearson correlation | Integration of RNA-seq and proteomics normalized log2-fold changes | Number of overlapping gene–protein pairs not stated | not stated |
| 2^(-ΔΔCt) comparative threshold cycle method (means ± SD reported; no formal significance test named) | RT-qPCR quantification of rpoS expression across OD600 growth-phase time points | 3 biological replicates each with 3 technical replicates | not stated |
| Gene Set Enrichment Analysis (GSEA) via fgsea R package | KEGG pathway enrichment for DEGs and DEPs | 928 DEGs and 832 DEPs used as ranked input lists | not stated |
| GO enrichment analysis filtered at raw p-value <0.05 | Gene Ontology (BP, MF, CC) enrichment for DEGs and DEPs | not stated | not stated |
-
Proteomics differential abundance was assessed using a nominal P-value threshold (<0.05) without a stated multiple-testing correction across thousands of proteins↳ Could also: Apply BH FDR correction at the protein level, as done for RNA-seq, or use a dedicated TMT-aware framework such as MSstats, DEqMS, or limma — FDR control across the full set of protein-level tests is standard practice in large-scale proteomics and would make the multiplicity treatment consistent with the RNA-seq analysis; MSstats and DEqMS are designed specifically for isobaric-label data and model within-protein variance explicitly
-
GO enrichment results were filtered at a raw p-value <0.05 across many simultaneously tested GO terms↳ Could also: Apply BH FDR correction to GO term p-values before filtering (e.g., q-value <0.05 or <0.1) — Testing hundreds of GO terms simultaneously inflates the chance of spurious enrichment calls; FDR-adjusted thresholds are widely used in gene-set analyses and reduce this risk while remaining less conservative than Bonferroni correction
-
Transcriptome–proteome correspondence was assessed by Pearson correlation of normalized log2-fold changes↳ Could also: Use Spearman rank correlation in addition to or instead of Pearson correlation — Spearman correlation requires no assumption of linearity or normality and is more robust to outliers, which are common in high-throughput omics fold-change distributions; reporting both would clarify whether any observed correlation is driven by extreme values
-
The growth-phase comparison used two discrete OD600 time points to represent exponential and stationary phases↳ Could also: Sample additional intermediate time points and model expression trajectories with time-series methods such as ImpulseDE2, maSigPro, or spline-based contrasts in DESeq2/edgeR — A time-series design captures kinetics of the exponential-to-stationary transition, enabling detection of genes with transient, delayed, or non-monotonic responses that a two-point comparison cannot distinguish
-
Dispersion around growth-curve and RT-qPCR means was reported as SD with n=3 biological replicates↳ Could also: Report 95% confidence intervals alongside SD, or report SEM with the n clearly stated — With small n, a 95% CI explicitly conveys precision around the mean estimate and scales with sample size, supporting inference about population-level differences; SD and CI serve complementary descriptive and inferential purposes and are often reported together
-
RT-qPCR data across OD600 time points were summarized as means ± SD without a named formal statistical comparison between growth phases↳ Could also: Apply a one-way ANOVA or mixed-effects model across OD time points with a post-hoc correction, or a two-sample t-test between the exponential and stationary time points used in the omics experiment — A formal test would provide a p-value to accompany the descriptive summary and would be consistent with the statistical thresholds applied in the RNA-seq and proteomics analyses
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 40680064
Title: Transcriptome-Proteome Profiling in Burkholderia thailandensis during the Transition from Exponential to Stationary Phase. J Proteome Res (2025), doi:10.1021/acs.jproteome.5c00223 · PMCID PMC12322963.
Code: https://github.com/Ahmed-Tohamy/rna_proteomics_integration (commit pinned at run time) RNA-Seq data: GEO GSE279483 (6 samples: 3 exponential, 3 stationary; HISAT2 + featureCounts + DESeq2) Proteomics data: ProteomeXchange/PRIDE PXD056625 (TMT; raw + mzid only — no per-protein summary table)
What the shipped repo actually does (the reproducible pipeline)
The repository is a single generic Python script (RNA-Seq_Proteomics_Integration.py,
~90 lines, pandas/numpy/scipy/sklearn/seaborn). It:
- reads two CSVs
rna_seq_data.csvandproteomics_data.csv, each withGeneID, log2FoldChange, p-value; - inner-merges them on
GeneID, drops NA; - z-score normalizes each
log2FoldChangecolumn withStandardScaler(note: an affine transform — does not change a Pearson r/p); - computes the Pearson correlation between the two log2FC vectors;
- emits a scatter+regression plot, a density plot, a violin plot, and
integrated_data.csv.
The repo does not ship the input CSVs nor the upstream RNA/proteomics DE pipelines. Per the study's P16 rule (third-party/own code is equally valid), the in-scope reproduction is running this integration logic on the paper's own deposited DE tables, plus recomputing the DE counts from those tables.
In scope (pipeline-derived, attempted)
| id | reported result | paper location | pipeline | how reproduced |
|---|---|---|---|---|
| C1 | Pearson r = 0.4, p = 8.63×10⁻¹⁰² between RNA-Seq & proteomics log2FC | Fig 6A + caption; "Integrative Analysis" | the repo script (pandas/scipy Pearson) | merge GEO RNA log2FC table × proteomics log2FC table on gene id, Pearson |
| C2 | 928 DEGs (564 up / 364 down), |log2FC|≥2 & FDR<0.05 | Results; Table S1 (si_002) | DESeq2 thresholding | recount from the deposited RNA DE table at the stated thresholds |
| C3 | 832 DEPs (552 up / 280 down), |log2FC|≥0.5 & p<0.05, of 3033 proteins | Results; Table S3 (si_004) | TMT DE thresholding | recount from the deposited protein table at the stated thresholds |
Out of scope / not attempted (and why)
- Full RNA-Seq alignment from FASTQ (HISAT2→featureCounts→DESeq2 from SRA raw reads): the hard 20%. The deposited GEO processed table already carries the per-gene DESeq2 log2FC/padj, so C1/C2 are reproducible from it without re-aligning. A from-FASTQ rerun is optional and only tests the upstream aligner/quantifier, which the repo does not ship. Skipped unless C2 fails to reconcile from the table.
- TMT quantification from raw MS (PXD056625 .raw/.mzid → protein log2FC): heavy proprietary-format reprocessing; the paper's protein-level log2FC table is the shipped derived product. Not attempted (out of 80/20 budget).
- KEGG/GO enrichment, STRING PPI networks, Cytoscape hub analysis (Tables S2/S4/S5, Figs): GUI / external-DB-dependent, non-deterministic against live databases; the repo does not implement them. Out of scope (non-pipeline).
- Wet-lab (growth curves, OD measurements, culture): out of scope.
Auditability flags to check at run time
- The reported p = 8.63×10⁻¹⁰² at r=0.4 implies a merged n of ≈2.4k gene/protein pairs (solved from the t-distribution). If the only obtainable protein table is the 832-DEP subset, the full-set r is not exactly reproducible and the implied n becomes the key audit signal — recorded explicitly, not asserted.
- StandardScaler before Pearson is a no-op for r; we report both raw and z-scored r to confirm the script's normalization does not (and cannot) alter the headline number.
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.
This is a strong reproduction with one localized, explainable deviation. The headline transcriptome-proteome correlation reproduces (r=0.389 → 0.4 at 1-dp; p astronomically significant, the order-of-magnitude offset fully explained by the unstated merged-n), and the RNA DEG counts reproduce exactly (928 = 564/364) from the authors' deposited DESeq2 table. The sole substantive gap is on the protein-selection side: the stated DEP rule yields 937 DEPs from the full table whereas 832 were published, with 105 threshold-passing 'High'-confidence proteins dropped by a criterion absent from Methods — an under-specification (possibly in SI), not fabrication, with direction (up>down) preserved. Net: central conclusion confirmed; deviation moderate and attributable to method under-specification / non-deposited merge cohort, so overall yellow.
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.