Academic Integrity Report: 'Isolation, Purification and Antioxidant Activity of Extracellular Polysaccharide from Candida guilliermondii'

Academic fraud report · Geng Detector

Summary

This report flags the 2016 paper published in Industrial Microbiology (DOI: 10.3969/j.issn.1001-6678.2016.02.005), authored by Sun Xiaomeng, Wang Shang, Cong Lina*, Li Cheng, and Dong Liang, as highly suspicious for data fabrication. Four substantive concerns were identified. First, the Q Sepharose F.F. linear elution profile (Figure 1) shows a NaCl concentration gradient (dashed line with triangle markers) rendered as a geometrically perfect straight line from the origin, which contradicts known liquid chromatography physics involving dead volume, pump mixing delay, and back-pressure effects. Second, the antioxidant assay (Figure 7-A) reports 100% hydroxyl radical scavenging at 0.36 mg/mL without error bars and with implausibly smooth dose-response curves, contrary to expected experimental noise from triplicate repeats. Third, the GC chromatogram (Figure 6) shows an unusually flat baseline with overly symmetric peaks, suggesting post-hoc noise removal or simulated data. Fourth, the species name 'Candida guilliermond' is misspelled throughout, missing the 'o' and altering the suffix. Verdict: highly suspicious, with strongest evidence relating to the elution curve and antioxidant data.

Verdict

🟠 Highly suspicious. Multiple independent anomalies point toward manipulated or fabricated figures rather than raw instrument output.

Key findings

Implausible linear gradient in ion-exchange chromatography (Figure 1): The NaCl concentration trace appears as a mathematically perfect straight line from zero, which is inconsistent with real HPLC/FPLC system behavior. Expected: a delayed, smooth S-shaped conductivity ramp.
Implausible 100% scavenging activity (Figure 7-A): The reported hydroxyl radical (·OH) clearance of 100% at 0.36 mg/mL is biochemically unrealistic; complete elimination of absorbance is rarely observed due to substrate limits, reaction kinetics, and instrument baseline.
Unnaturally clean GC chromatogram (Figure 6): Baseline is exceptionally flat in the 17–21 min window; peaks are overly sharp and symmetric, with no detector noise or column bleed visible, suggesting noise suppression or simulated plotting.
Species name misspelled: The yeast is consistently written as *Candida guilliermond* (missing an 'o' and altered suffix) instead of the accepted *Candida guilliermondii* / *Meyerozyma guilliermondii*.

Evidence highlights

Figure 1, Page 3: NaCl gradient shown as a perfect straight dashed line with triangle markers beginning at 0 — no S-curve delay, no mixing artifact.
Figure 7-A, Page 4: Dose-response curve for EPS2-1 hydroxyl radical scavenging; reaches the 100% plateau at 0.36 mg/mL with no error bars; Figures 2, 3, 4 and 7 all lack visible replicate noise.
Figure 6, Page 4: GC trace of polysaccharide hydrolysate — baseline flat, peaks near-perfect Gaussian, no thermal drift visible across the 17–21 min retention window.
Abstract, Section 1.1.1, and English abstract: persistent misspelling *Candida guilliermond*.
DOI: 10.3969/j.issn.1001-6678.2016.02.005 — Vol. 46, No. 2, April 2016.

Notes

All four findings were internally consistent and reviewed as confirmed by the analyst.
The strongest single piece of evidence is the perfectly linear salt gradient in Figure 1, which is physically very difficult to obtain from real chromatography hardware.
Recommended follow-up: request raw instrument export files (e.g., .cdf or chromatography workstation logs) and raw absorbance readings from the antioxidant assays; consider posting the concerns on PubPeer; cross-check the authors' other publications for similar 'overly perfect' data patterns.
This report is AI-assisted and intended for academic discussion only; final determination of misconduct requires an official institutional investigation.