dnoise: Fast Native Data Reduction for Bruker timsTOF
Citation integrity
Citation Integrity Audit Report
Manuscript: dnoise: Fast Native Data Reduction for Bruker timsTOF
Checklist Categories in Play
The manuscript contains:
- Reference resolvability triggers: Specific citations to prior work, software, databases, and data repositories throughout
- Claim–citation support triggers: Quantitative claims about instrument capabilities, prior tools, and benchmark data attributed to references
- Quotation/number fidelity triggers: Specific values and ratios cited from benchmark sources
- Self-citation / citation inflation: Present but appears germane (authors cite their own software release and benchmark participation)
- Retracted / predatory sources: No obvious flags, but verification needed for recent works
All three categories require checking.
Findings by Category
1. Reference Resolvability
Load-bearing references checked:
| Ref | Citation | Status | Notes |
|---|---|---|---|
| (1) | Fernandez-Lima et al. 2011, IJIMS 14(2–3):93–98, DOI 10.1007/s12127-011-0067-8 | PRESENT | Foundational TIMS instrumentation; DOI resolvable |
| (2) | Cumeras et al. 2015, Analyst 140(5):1376–1390, DOI 10.1039/C4AN01100G | PRESENT | Ion mobility review; DOI resolvable |
| (3) | Meier et al. 2015, J. Proteome Res. 14(12):5378–5387, DOI 10.1021/acs.jproteome.5b00932 | PRESENT | PASEF method; DOI resolvable |
| (4) | Houthuijs et al. 2026, Anal. Chem. 98(20):15066–15074, DOI 10.1021/acs.analchem.6c00762 | UNVERIFIABLE | Future-dated (August 2026 manuscript, reference dated 2026); DOI format appears valid but publication status cannot be confirmed. This is a SOFT concern—likely a preprint or advance online publication, but requires author clarification. |
| (5) | Martens et al. 2011, mzML standard, Mol. Cell. Proteomics 10(1):R110.000133, DOI 10.1074/mcp.R110.000133 | PRESENT | Standard format reference; DOI resolvable |
| (6) | Pfeuffer et al. 2024, OpenMS 3, Nat. Methods 21(3):365–367, DOI 10.1038/s41592-024-02197-7 | PRESENT | Recent software; DOI resolvable |
| (7) | Bilbao et al. 2022, PNNL PreProcessor, J. Proteome Res. 21(3):798–807, DOI 10.1021/acs.jproteome.1c00425 | PRESENT | Cited as prior denoising tool for ion mobility; DOI resolvable |
| (8) | Prianichnikov et al. 2020, MaxQuant, Mol. Cell. Proteomics 19(6):1058–1069, DOI 10.1074/mcp.TIR119.001720 | PRESENT | Ion mobility software; DOI resolvable |
| (9) | Yu et al. 2020, MSFragger/IonQuant, Mol. Cell. Proteomics 19(9):1575–1585, DOI 10.1074/mcp.TIR120.002048 | PRESENT | Quantification tool; DOI resolvable |
| (10) | Łącki et al. 2021, OpenTIMS/TimsPy/TimsR, J. Proteome Res. 20(4):2122–2129, DOI 10.1021/acs.jproteome.0c00962 | PRESENT | Data access libraries; DOI resolvable |
| (11) | Willems et al. 2021, AlphaTims, Mol. Cell. Proteomics 20:100149, DOI 10.1016/j.mcpro.2021.100149 | PRESENT | Visualization tool; DOI resolvable |
| (12) | Langella et al. 2024, i2MassChroQ, J. Proteome Res. 23(8):3353–3366, DOI 10.1021/acs.jproteome.3c00732 | PRESENT | Native timsTOF processing; DOI resolvable |
| (13) | Teschner et al. 2025, Rustims, J. Proteome Res. 24(5):2358–2368, DOI 10.1021/acs.jproteome.4c00966 | PRESENT | Rust framework for timsTOF; DOI resolvable |
| (14) | Wilding-McBride et al. 2022, spectral simplification, PLoS One 17(7):e0271025, DOI 10.1371/journal.pone.0271025 | PRESENT | Cited as prior denoising method; DOI resolvable |
| (15) | Willems & MannLabs, timsrust 0.4.2, GitHub link provided | PRESENT | Software library; GitHub URL provided (https://github.com/MannLabs/timsrust) |
| (16) | Deutsch et al. 2020, ProteomeXchange, Nucleic Acids Res. 48(D1):D1145–D1152, DOI 10.1093/nar/gkz984 | PRESENT | Data repository standard; DOI resolvable |
| (17) | Perez-Riverol et al. 2022, PRIDE Database, Nucleic Acids Res. 50(D1):D543–D552, DOI 10.1093/nar/gkab1038 | PRESENT | Data repository; DOI resolvable |
| (18) | Van Puyvelde & Dhaenens, LFQ Benchmark Generation Beta, DOI 10.6019/PXD070049 | PRESENT | Benchmark dataset; DOI resolvable (PRIDE accession) |
| (19) | Van Puyvelde et al., same benchmark, DOI 10.64898/2026.01.29.702266 | UNVERIFIABLE | DOI format appears malformed (10.64898 is not a standard DOI prefix; standard prefixes are 10.xxxx where xxxx is 4+ digits). The date string in the DOI (2026.01.29) is unusual. This may be a transcription error. SOFT issue—the benchmark is clearly the same as ref. 18, but this alternate citation needs clarification. |
| (20) | Navarro et al. 2016, LFQ benchmark multicenter, Nat. Biotechnol. 34(11):1130–1136, DOI 10.1038/nbt.3685 | PRESENT | Quantification benchmark; DOI resolvable |
| (21) | Lazear 2023, Sage search engine, J. Proteome Res. 22(11):3652–3659, DOI 10.1021/acs.jproteome.3c00486 | PRESENT | Search tool used in benchmark; DOI resolvable |
| (22) | Yu et al. 2021, IonQuant, Mol. Cell. Proteomics 20:100077, DOI 10.1016/j.mcpro.2021.100077 | PRESENT | Quantification method; DOI resolvable |
| (23) | Demichev et al. 2020, DIA-NN, Nat. Methods 17(1):41–44, DOI 10.1038/s41592-019-0638-x | PRESENT | DIA search tool used in benchmark; DOI resolvable |
| (24) | Cox et al. 2014, MaxLFQ, Mol. Cell. Proteomics 13(9):2513–2526, DOI 10.1074/mcp.M113.031591 | PRESENT | Quantification method; DOI resolvable |
| (25) | Elias & Gygi 2007, target-decoy search, Nat. Methods 4(3):207–214, DOI 10.1038/nmeth1019 | PRESENT | Statistical method; DOI resolvable |
Summary for resolvability:
- 25 references total
- 22 PRESENT (resolvable with valid DOI/URL)
- 2 UNVERIFIABLE (refs. 4, 19 — future date or malformed DOI)
- 0 MISSING
2. Claim–Citation Support
Key factual claims checked against cited references:
Claim 1 (Introduction, lines 50–53): "On other ion-mobility platforms, the PNNL PreProcessor writes denoised data back in the instrument's own format, but it does not support Bruker .d."
- Citation: Ref. (7) Bilbao et al. 2022
- Status: UNVERIFIABLE — The reference is to a preprocessing tool for ion-mobility MS workflows, but the manuscript does not quote or provide the abstract/methods section. The claim that it writes in native format for non-Bruker platforms is plausible but cannot be confirmed from the manuscript alone. Recommend author provide excerpt or clarification.
- Severity: SOFT (not a central claim; contextual positioning of prior work)
Claim 2 (Introduction, lines 56–57): "In representative 5-minute runs, 50.1% of ddaPASEF and 72.7% of diaPASEF MS1 points lay outside the fragmentable region."
- Citation: None explicitly given; appears to be from the authors' own data
- Status: PRESENT — This is the authors' own observation from their benchmark, not attributed to prior work. No citation required.
Claim 3 (Methods 2.5, lines 132–143): "We used the Generation Beta three-species hybrid benchmark (human, Saccharomyces cerevisiae, and Escherichia coli) deposited as PRIDE PXD070049. Its defined ratios provide a standard test of label-free quantification accuracy."
- Citation: Refs. (18) Van Puyvelde & Dhaenens and (20) Navarro et al. 2016
- Status: PRESENT — Ref. 18 is the benchmark dataset itself (PRIDE PXD070049); Ref. 20 is cited for the LFQ benchmark methodology. Both are appropriate and resolvable.
Claim 4 (Methods 2.6, lines 145–159): "Sage provided mobility-aware MS1 label-free quantification... identifications were transferred between runs using decoy-controlled LFQ q-values, as in IonQuant."
- Citation: Refs. (21) Lazear 2023 (Sage) and (22) Yu et al. 2021 (IonQuant)
- Status: PRESENT — Both tools are cited for their respective methods. Appropriate.
Claim 5 (Methods 2.6, lines 160–162): "The three diaPASEF arms were searched independently with DIA-NN 2.2.0, which read each native .d directory through its bundled Bruker timsdata library."
- Citation: Ref. (23) Demichev et al. 2020
- Status: PRESENT — DIA-NN is cited; the claim about bundled Bruker support is plausible for a 2020 tool but not explicitly verified in the manuscript excerpt. However, this is a factual claim about software capability, not a quantitative result, and the tool is correctly cited.
Claim 6 (Methods 2.6, lines 163–164): "DIA-NN quantified proteins with MaxLFQ from MS2 fragment chromatograms."
- Citation: Ref. (24) Cox et al. 2014
- Status: PRESENT — MaxLFQ is correctly cited as the quantification method.
Claim 7 (Methods 2.7, lines 167–168): "Accuracy is the species-specific median log₂ ratio relative to the known mixture ratio."
- Citation: Implicit reference to benchmark design (refs. 18, 20)
- Status: PRESENT — This is a standard metric for the benchmark; no new claim.
Claim 8 (Results 3.4, lines 260–276): "A rank-1 decoy hit is a match to a sequence known to be absent from the sample... In the original ddaPASEF searches at 5 and 15 minutes, respectively, 30.1% and 23.7% of scored spectra had a decoy as their best available explanation."
- Citation: Ref. (25) Elias & Gygi 2007 (target-decoy search method)
- Status: PRESENT — The target-decoy method is correctly cited. The percentages are the authors' own calculations from their benchmark data, not attributed to the reference.
3. Quotation and Number Fidelity
No direct quotations from prior work are provided in the manuscript. Numerical claims are either:
- The authors' own benchmark results (e.g., "50.1% of ddaPASEF MS1 points"), or
- Standard methodological parameters (e.g., "±20 ppm precursor and fragment tolerances")
Status: PRESENT — No quotation fidelity issues detected.
4. Self-Citation and Citation Inflation
Self-citations identified:
- Ref. (15): timsrust library (MannLabs contributors) — not authored by Garrett et al., but used by them
- Ref. (18): LFQ Benchmark Generation Beta — Van Puyvelde & Dhaenels are the benchmark creators; Garrett et al. are users/contributors
- Software release: dnoise v0.1.0 (Zenodo DOI 10.5281/zenodo.21959649) — the authors' own tool, appropriately cited in Data Availability
Assessment: The self-citations are germane and appropriate. The authors cite the benchmark they used (which they may have contributed to) and their own software release (required for reproducibility). No citation inflation detected.
Severity: SOFT (not a concern; standard practice)
5. Retracted or Predatory Sources
All cited journals are mainstream, peer-reviewed venues:
- Nature Methods, Nature Biotechnol., Nucleic Acids Res., Mol. Cell. Proteomics, J. Proteome Res., Analyst, PLoS One, Anal. Chem., Int. J. Ion Mobility Spectrom.
Status: PRESENT — No retracted or predatory sources detected.
Summary Table
| Category | Trigger Present | Findings | Severity |
|---|---|---|---|
| Reference Resolvability | Yes | 22/25 present; 2 unverifiable (refs. 4, 19); 0 missing | SOFT (refs. 4, 19) |
| Claim–Citation Support | Yes | All load-bearing claims supported or self-generated; 1 unverifiable detail (ref. 7 scope) | SOFT (ref. 7) |
| Quotation/Number Fidelity | No | No direct quotations; no fidelity issues | — |
| Self-Citation Inflation | Yes | Appropriate and germane; no inflation | — |
| Retracted/Predatory Sources | No | No problematic sources | — |
Issues Requiring Author Clarification
Issue 1: Reference 4 (Houthuijs et al. 2026) — Future-dated publication
Finding: The manuscript is dated August 2026, and this reference is also dated 2026. The DOI format is valid, but the publication status cannot be independently confirmed.
Question for authors: Is this an advance online publication, preprint, or in-press article? Please provide the publication status and confirm the DOI is correct.
Severity: SOFT
Issue 2: Reference 19 (Van Puyvelde et al.) — Malformed DOI
Finding: The DOI is listed as 10.64898/2026.01.29.702266, which has an unusual structure (date string embedded in DOI). This does not match standard DOI formatting.
Question for authors: Is this DOI correct? Reference 18 cites the same benchmark with DOI 10.6019/PXD070049 (a PRIDE accession). Please clarify whether ref. 19 is a duplicate, a different version, or a transcription error.
Severity: SOFT
Issue 3: Reference 7 (Bilbao et al. 2022) — Scope of prior tool
Finding: The manuscript claims the PNNL PreProcessor "writes denoised data back in the instrument's own format" for non-Bruker platforms but does not support Bruker. The cited reference is resolvable, but the manuscript does not provide evidence that this claim is explicitly stated in the paper.
Question for authors: Can you provide a quote or section reference from Bilbao et al. 2022 confirming that the PreProcessor supports native-format output for non-Bruker platforms and explicitly does not support Bruker .d?
Severity: SOFT (contextual claim, not central to the paper's contribution)
Conclusion
Overall status: COMPLIANT with minor clarifications needed
- All 25 references are resolvable or present in the reference list.
- Two references (4, 19) require author clarification on publication status and DOI format.
- All load-bearing claims are either supported by cited references or are the authors' own data.
- No retracted, predatory, or obviously false citations detected.
- Self-citations are appropriate and germane.
Recommendation: Request author responses to the three clarification questions above before final acceptance. None of these issues block reproducibility or invalidate the central claims.