Accuracy, precision and the tolerance nobody states
Parts per million sound impressive until they are converted into daltons at the molecular weight of the thing being measured.
Analytics
Page 11 of 25 of this archive, newest first.
Page 11
Independent analysis puts a QST orforglipron lot at 98.0%, against 94.2% on the certificate
The report states the gradient, the wavelength and the integration threshold, which is more than most.
Twenty-eight days is a number from somebody else’s product
An in-use period is established by a dedicated study on a specific formulation in a specific container at a specific concentration. Borrowing one from a marketed product is…
One method is one opinion
Where two methods disagree, the conservative convention is to report the lower figure. It is not universal, and whether a laboratory follows it belongs on the report.
VendorInvestigate result on a Changsha liraglutide lot lands 3.2 points below the supplier’s figure
A reminder that a purity figure is the output of a method, and that methods differ.
PeptideMeter result on a Hong Kong mazdutide lot lands 1.4 points below the supplier’s figure
The Journal submitted the sample and paid for the analysis. The vendor was told in advance.
Independent analysis puts a SGN semaglutide lot at 96.6%, against 94.9% on the certificate
The supplier has not disputed the finding. It has not explained the gap either.
Independent analysis puts a JEEP retatrutide lot at 96.6%, against 92.9% on the certificate
The report states the gradient, the wavelength and the integration threshold, which is more than most.
Resolving power and mass accuracy are two different specifications
Parts per million sound impressive until they are converted into daltons at the molecular weight of the thing being measured.
VendorInvestigate result on a Shenzhen semaglutide lot lands 1.1 points below the supplier’s figure
The result is unremarkable. What the report omits is not.
Same mass, different molecule: why identity by mass is not identity
Reversed-phase chromatography runs in an organic, acidic mobile phase that dissociates most non-covalent aggregates on the way to the detector.
Independent analysis puts a GGPeps liraglutide lot at 97.8%, against 93.8% on the certificate
The Journal submitted the sample and paid for the analysis. The vendor was told in advance.
Medutest result on a Xi'an orforglipron lot lands 1.0 points below the supplier’s figure
The report states the gradient, the wavelength and the integration threshold, which is more than most.
Date of analysis, date of manufacture, and the gap between them
The interval between manufacture and analysis is the most under-read figure on the page, and the one most likely to matter by the time a vial is opened.
Independent analysis puts a MKM dulaglutide lot at 99.2%, against 96.7% on the certificate
The report states the gradient, the wavelength and the integration threshold, which is more than most.
Janoshik result on a Nantong mazdutide lot lands 0.9 points below the supplier’s figure
The result is unremarkable. What the report omits is not.
Janoshik result on a Wuxi semaglutide lot lands 2.4 points below the supplier’s figure
The supplier has not disputed the finding. It has not explained the gap either.
Independent analysis puts a SGN orforglipron lot at 96.1%, against 93.2% on the certificate
The result is unremarkable. What the report omits is not.
What we asked twenty suppliers about stability data
A retest date says that material should be re-examined before use. An expiry date says it should not be used. The trade uses the second word for the first concept.
Deamidation, oxidation, and the two shifts worth memorising
What would be needed to catch each of these, and what it would cost.