Integrity audit of Nature Communications paper 10.1038/s41467-022-29270-z: 'Mechanism of action and therapeutic route for a muscular dystrophy caused by a genetic defect in lipid metabolism' (Tavasoli, McMaster et al., 2022)

Verdict

🟠 Highly suspicious. Six textual inconsistencies were identified; five are internally verified from the manuscript text. The authors have not yet been contacted. The findings are consistent with careless drafting or template reuse, but they also raise the possibility of data fabrication. No image-level forensic audit has been performed.

Key findings

• Sampling-week mismatch (Results vs Methods): Figure 1 legend and Results state serum CK was measured at weeks 6, 12, and 17, while Methods (Creatine kinase serum levels) states blood was collected at 5, 10, and 15 weeks of age.
• Non-existent gene probe: The Methods section lists a TaqMan probe called 'Chkb Cpt1b' (Exon boundary 7-8) with RRID Mm01308102_g1. No such single gene exists; this appears to be a pasted concatenation of 'Chkb' and 'Cpt1b'. A separate Cpt1b probe (Mm00487191_g1) is also listed elsewhere.
• Genotyping label error: The Results text states histology was performed on 'Chkb^{+/+}, Chkb^{+/-} and Chkb^{+/-}' mice, with Chkb^{-/-} missing and the heterozygote listed twice.
• Misattributed figure callout: The text cites Fig. 1F for hindlimb muscle weights (EDL, gastrocnemius, quadriceps, TA), while the Fig. 1 legend identifies 1F as serum creatine kinase; the bar chart for muscle weights is labelled 1G.
• Malformed P-values: Fig. 3 legend contains 'p=0.0.2500' and 'p=0.0.0001', which are not valid outputs from any standard statistics package and suggest manual editing or a transcription error.
• Image audit not feasible: The available content is text-only; Western blots in Figs. 2 and 4 and other panels could not be examined for splicing or duplication.

Evidence highlights

• Finding 1 (weeks): Results section / Fig. 1F caption: 'weeks 6, 12, and 17'. Methods (Creatine kinase serum levels): '5, 10, and 15 weeks old'. Offsets of 1–2 weeks on every time point.
• Finding 2 (probe): Methods qPCR paragraph: 'Chkb Cpt1b (Exon boundary7-8)(RRID: Mm01308102_g1)'. A separate Cpt1b probe (Mm00487191_g1) is listed in the same paragraph.
• Finding 3 (genotypes): Results paragraph: '25-day old Chkb^{+/+}, Chkb^{+/-} and Chkb^{+/-} mice (Supplementary Fig. 2A-C).' Chkb^{-/-} is absent.
• Finding 4 (figure): Results text cites 'Fig. 1F' for muscle weights; Fig. 1 legend labels 1F as serum CK and 1G as normalised muscle weights.
• Finding 5 (P-values): Fig. 3 legend contains the strings 'p=0.0.2500' and 'p=0.0.0001'.
• Finding 6 (images): Not analysed; flagged for follow-up with the authors and editorial office.

Notes

• Limitations: detection is text-based; no pixel-level, statistical re-analysis, or raw-data check has been performed. Conclusions rest on internal inconsistencies in the published manuscript and on the absence of independent verification rather than on proven data fabrication.
• Possible benign explanations include template reuse, copy-paste errors, and figure renumbering during revision. However, the combination of a fabricated gene name, an irreconcilable timeline, and malformed P-values lowers the prior probability of a purely innocent explanation.
• Recommended actions: (1) request raw qPCR data, CK measurement records with exact sampling dates, and uncropped original blots from the corresponding author; (2) raise a PubPeer comment enumerating findings 1–5; (3) notify the Nature Communications editorial office regarding the methodological inaccuracies; (4) consider an institutional complaint only after the authors' response.
• DOI preserved from the original report: 10.1038/s41467-022-29270-z.
• This report is AI-assisted and intended for academic discussion only; any final determination of misconduct requires an institutional investigation.