leena
Dear all,
I have a question regarding the fitting of Example 2 (T3R3 human insulin Zn complex). It says in the tutorial that first 24 peaks can be easily fitted as FP-type with the following parameters: lambda = 1.4011, Simple Axial model, Cry Size L. However it doesn't result in a good fit at all. Setting the Cry Size L parameters to the same value, changing the number of background parameters, fitting the peaks one by one could not help. The main problem is that the peaks move far away from their expected positions and the fit fails. Constraining the peak positions does not solve the problem either.
I would appreciate any suggestion or hint.
Leena
johnsoevans
Which tutorial are you talking about? Can you post a link?
leena
johnsoevans
Dear Leena, I'm afraid I've not done that tutorial as it's one of the official Bruker ones. However knowing how good their support material is I'm surprised that it doesn't work. From the image in the pdf it does look to me as if the simple peak shape model they suggest should work. I'd suggest you initially try fixing peak positions and refining just the asymmetry and a single CS_L parameter. If that looks ok then free the peak positions.
If you have a commercial version of topas you might be able to get help directly from them.
leena
Thank you very much for your suggestion. I can get a proper fitting if I constrain peak positions around their maxima and CS_L parameter to be 200. Then I can refine CS_L (set to the same value for all peaks) and peak positions.
henrik
This is one of the trickier patterns and the tutorial is removed from the later TOPAS versions. Before refining the positions, refine the Parameter for the simple axial model using the first 10 peaks, keeping the positions fixed. It should end up at around 5.2mm. After that you can add peaks and refine them, it makes the refinement more stable.