Different error values during Guinier analysis

Interactive and automated data processing and analysis tools: PRIMUS, CHROMIXS, GNOM, AUTORG, DATCMP, DATOP etc.
Post Reply
Posts: 1
Joined: 2016.11.11 14:45

Different error values during Guinier analysis

#1 Post by Leanthil » 2016.11.11 15:35

Dear all,

I have a problem:

In primusqt (ATSAS version 2.7.2) during "Radius of Gyration" analysis I get a suggestion for Rg with its error automatically as expected. However, when I alter the data ranges the error suddenly drops almost an order of magnitude. (E.g. from 4.53+-0.27 to 4.53+-0.04). If I change the data ranges back to its original (as autorg suggested) the errors do not return to their original values.
Similar happens with the I0 estimation as well, however the changes are not this dramatic.

I've tried multiple datasets of various quality and source (P12, SWING), but this behavior is persistent.

Based on the quality of my data I would trust the initial error estimation, however in cases when I have to alter the ranges (e.g. aggregation), the errors become surprisingly low. Which error value should I trust, or what could I do in these cases to get a more reasonable error estimate?

Thank you.


User avatar
Posts: 717
Joined: 2007.08.03 18:55
Location: EMBL Hamburg, Germany


#2 Post by AL » 2016.11.18 12:45

When you open the Guinier analysis window ("Radius of Gyration" button in PRIMUS) the AUTORG program is executed behind the scenes. AUTORG automatically finds the best Guinier interval and prints (among other values) the standard deviation of the experimental data from the Guinier fit in the selected interval plus standard deviation of Rg values from all possible intervals from the Rg value from the selected interval. This way the accuracy of Rg is estimated by taking into account not only the error propagation in the selected fit but also by accounting for the deviation of Rg values calculated from other consistent intervals, accounting to some extent for systematic errors in the Rg determination.

When you change the automatically suggested "Range" the program DATRG is executed to compute the Rg from the given interval. DATRG computes the error only based on the deviation of the experimental data from the Guinier fit in the selected range without taking other possible ranges into account. Thus this error is underestimated.

Post Reply