Review Papers Claim by Claim

 Long Explanation Copied

What the paper actually did

Tang, Ye, Feng and Quinn (Nat. Prod. Rep., 2015) reviewed 84 major/bioactive compounds from 15 TCM herbs traditionally used for neurodegenerative symptoms, computed six physicochemical descriptors (MW, log P, HBA, HBD, ROTB, PSA) in Instant JChem, compared them against 38 anti-PD and 65 anti-AD drugs/candidates from Thomson Reuters Cortellis, ran ChemGPS-NP PCA, and predicted BBB permeation with QikProp (QPlogBB, QPPCaco, QPPMDCK) Original TCM drug-space review.

Key quantitative findings (recomputed from Table 2–4 data)

The stacked bars verify the reported 92% compliance (\u22641 violation) for anti-PD and anti-AD sets versus 71% for TCM compounds Ro5 compliance statistics. Table 4 data confirm the authors' mechanistic attribution: 22 of the 24 Ro5-failing TCM compounds (e.g., ginsenosides 1\u20139, rehmanniosides, jujubosides) carry at least one sugar moiety inflating HBA/HBD/MW \u2014 a structural explanation, not just a statistical pattern.

BBB predictions and the circularity problem

Using QikProp, 54.8% of the 84 compounds gave both ideal QPlogBB (\u22123.0 to 1.2) and QPPCaco/QPPMDCK >25; 23.8% showed good predicted permeability, which the authors call \u201cimpressive\u201d against Pardridge's claim that >98% of small-molecule drugs do not cross the BBB In silico BBB permeability results. BGPT flags a hidden assumption here: the herbs were selected because they are traditionally used for brain symptoms, so finding brain-penetrant-like properties is partly preordained by selection \u2014 the 23.8% figure supports enrichment, not evidence of clinical efficacy. The in silico predictions also rely on QikProp heuristics (2009 version), not measured brain concentrations, though the authors do cite consistent in vitro/in vivo data for a subset (tanshinones, asarones, senkyunolide I).

Critical appraisal

• Strength: the ChemGPS-NP PCA showing most TCM compounds falling within the drug-like space occupied by launched CNS drugs is a legitimate, reproducible cheminformatic comparison; the descriptor tables are fully reported, enabling independent verification.
• Weakness \u2014 conflating drug-likeness with drug potential: physicochemical overlap says nothing about potency, target engagement, or efficacy. The narrative repeatedly notes neuroprotective effects from in vitro PC12/SH-SY5Y assays and mouse models, but no human clinical evidence for any of these 84 compounds in AD/PD is presented.
• Weakness \u2014 saponin dismissal is premature: the authors note saponins fail Ro5 but have \u201cfavourable ADME/T properties\u201d citing GM1 ganglioside, then leave the class \u201crequiring further investigation\u201d \u2014 internally inconsistent framing without proposing how metabolites (e.g., Compound K from ginsenosides) might be the true active species, which their own pharmacokinetic citations support.
• Sponsor context: funded by China's \u201cKey New Drug Creation\u201d national program and Griffith scholarships; no conflict-of-interest statement is provided. Positive framing toward TCM drug-discovery potential warrants awareness of funder incentive Funding disclosure.
• Dated evidence base: covers only literature to 2015; Cortellis drug lists and QikProp models have been superseded. What would falsify the central claim: RCTs of the highlighted BBB-permeant compounds (e.g., ligustrazine, tanshinone IIA) showing no clinical benefit in AD/PD patients despite the favorable drug-space alignment.

Author reviews

 BGPT Paper Review

Study Novelty

Applies established cheminformatics tools (Ro5, ChemGPS-NP, QikProp) to TCM natural products for CNS indications \u2014 a sensible but not novel methodological transfer; prior analyses of natural product drug-likeness existed (Quinn 2008, Feher & Schmidt 2003).

Scientific Quality

Transparent descriptor tables and PCA enable verification; however, selection circularity, absence of efficacy evidence, lack of COI statement, and reliance on 2009-era in silico BBB heuristics temper quality.

Study Generality

The drug-space comparison framework generalizes to any ethnomedical compound collection, though the specific compound list is AD/PD-focused.

Study Usefulness

Provides a prioritized, BBB-filtered shortlist of TCM lead candidates (lactones, diterpene quinones, sesquiterpenes) directly usable by medicinal chemists; limited by absence of potency/target data.

Study Reproducibility

All 187 compound descriptor values are tabulated; tools (Instant JChem, ChemGPS-NP, QikProp) are named with versions, enabling independent re-analysis.

Explanatory Depth

Explains Ro5 violations structurally (sugar moieties) but offers no mechanistic account of how the compounds act on neurodegenerative pathology beyond citing heterogeneous in vitro studies.