Geng Academic Fraud Detection Report — Butyrate Suppresses Glucose Metabolism of Colorectal Cancer Cells via GPR109a-AKT Signaling Pathway and Enhances Chemotherapy (Frontiers in Molecular Biosciences, 2021)

Academic fraud report · Geng Detector

Summary

Verdict: Highly suspicious. The report identifies multiple irregularities in the paper by Geng et al. (DOI: 10.3389/fmolb.2021.634874) that collectively suggest data fabrication or copy-paste methodology rather than minor sloppiness. Key concerns: (1) Reported apoptosis rates (9% butyrate alone, 13% 5-FU alone, 22% combination) follow an implausibly perfect additive relationship (9+13=22), inconsistent with expected biological variance. (2) The CCK-8 viability assay is described with an absorbance of 590 nm, which corresponds to MTT/BCA protocols; CCK-8 (WST-8) is read at ~450 nm. (3) Reported metabolite reductions (70%, 58%, 60%) are suspiciously round integers for LC-MS/MS-class measurements. (4) The Western Blot buffer recipe contains apparent PDF line numbers ('274', '275'), indicating copy-paste from another manuscript. (5) The paper forcefully inserts COVID-19 relevance into an in vitro colorectal cancer study. No pixel-level image analysis was possible from the report. Confidence: moderate-to-high for methodological issues; high uncertainty on quantitative data fabrication without raw flow cytometry files.

Verdict

🟠 Highly suspicious. Multiple independent irregularities (mathematically perfect numbers, a wavelength error inconsistent with the stated kit, copy-paste artifacts in Methods, opportunistic COVID-19 framing) collectively exceed what could be attributed to ordinary error. They are consistent with templated, low-integrity manuscript production. No image-level pixel analysis was performed in this report.

Key findings

Perfect-additive apoptosis rates: Butyrate 9% + 5-FU 13% reported as yielding exactly 22% combined apoptosis — an implausible, noiseless additivity from flow cytometry.
Wrong absorbance wavelength for CCK-8: Methods describe reading CCK-8 (WST-8) at 590 nm; the assay's standard read is ~450 nm. 590 nm is typical for MTT or BCA assays.
Round-integer metabolite reductions: R5P, acetyl-CoA, and NADPH reportedly fall by exactly 70%, 58%, and 60% — unusually clean for metabolomics.
Copy-paste artifacts in RIPA buffer recipe: Stray tokens "274" and "275" in the Western Blot method are almost certainly ingested PDF line numbers.
Forced COVID-19 linkage: A purely in vitro colorectal cancer paper repeatedly invokes the COVID-19 pandemic despite no viral/antiviral experimental component.
Image analysis not performed: The report explicitly states pixel-level inspection of Figures 1–5 was not possible from the available inputs.

Evidence highlights

Apoptosis numerics from the Results section describing Figure 5: butyrate 9%, 5-FU 13%, combination 22% (9 + 13 = 22).
Methods — Cell Viability Assay: "Absorbance at 590 nm" paired with Cell Counting Kit-8 (CCK-8); CCK-8/WST-8 absorbance maximum is ~450 nm.
Methods — Western Blotting: "1% NP-40, 0.5% deoxycholic acid, 274 and 0.1% SDS"; "(1 mM PMSF, 5 μg/ml 275 leupeptin…".
Results — Butyrate Inhibited G6PD Expression and DNA Synthesis (Supplementary Figure S4): metabolite reductions "by 70, 58, and 60%, respectively."
Introduction and Conclusion: extensive COVID-19 discussion with no corresponding experimental design.

Notes

Confidence is moderate-to-high for methodological anomalies (wavelength error, copy-paste line numbers) because they are objectively inconsistent with the stated protocols.
Confidence on quantitative fabrication (the 9+13=22 case, the round percentages) is suggestive but not conclusive without raw flow cytometry data and the original metabolomics output; alternative explanations (extreme rounding during figure preparation, mis-typed numbers) cannot be fully excluded.
All numeric quotations and the DOI (10.3389/fmolb.2021.634874) are preserved verbatim from the source report.
Recommended next actions listed in the source report (request raw FCS files, PubPeer post, formal erratum inquiry) remain appropriate. Journal-level investigation by Frontiers in Molecular Biosciences should be considered if independent parties can supply raw data.