Dear Daniel, A very good question. I have seen R_d plots jumping up and down, most often in low-symmetry space groups. I think the "mountains" are due to residual systematic differences between symmetry-related observations, differences that the scaling could not compensate for whatever reason. They are most likely associated with observations on _different_ quadrants of the detector - it is on my list of things to track down.
Summary: this tells you nothing about radiation damage. Radiation damage can be concluded from the overall slope of R_d, not from local features. Hope this makes sense, Kay On Thu, 17 Oct 2013 17:48:45 +0200, Danilo Belviso <[email protected]> wrote: > Dear all, > > I am using xds (with graphical interface xdsgui) to process several > diffraction data of a membrane protein that I have crystallized. At the > end, I run XDSSTAT in order to check the statistic parameters of the > process and my attention is captured by the R_d plot: R_d drops during > the firsts 10-15 frames and then reaches a maximum (20-30 frames). Then, > its value remains quite stable (I suppose due to the radiation damage > correction performed by xds). The trend seems the same that is shown in > > ActaCryst. (2006). D62, 96–101 where "Some aspects of quantitative > analysis and correction of radiation damage" > Fig.1(a)(b) > > where R_d (decay R-factor) was introduced. > > The question is: > > By considering that radiation damage increases during the data > collection due to the progressive dose of radiation to which the crystal > is subjected, why does the R_d drop during the first period of data > collection? > > Thank you in advance for your answers. > > Danilo
