I also think Rpim, as commonly defined, is not directly useful for measuring 
the anomalous signal.
Commenting on its use, though, I find it quite useful to judge whether one is 
adding useful data
to already existing ones. In a multiple crystals context I always look at both 
Rmeas and Rpim.
Suppose a given dataset shows some Rmeas and Rpim. Now add another dataset to 
the first one, in
the hope to increase completeness and redundancy. This, of course, will depend 
on how isomorphic
were the crystals from which the two datasets were collected. If the Rmeas 
stays stationary, or increases 

just a little, and Rpim decreases, this means thatthe crystals had a good 
degree of isomorphism and 

that scaled data will be more precise. This is what one wished for, when 
collecting from multiple crystals.
On the other hand, an increase in Rpim definitely points at something wrong 
with the addition of the
second dataset, quite certainly some form of non-isomorphism.


Dr James Foadi PhD
Membrane Protein Laboratory
Diamond Light Source Ltd.
Diamond House
Harwell Science and Innovation Campus
OX11 0DE
United Kingdom

office email: james.fo...@diamond.ac.uk
alternative email: j.fo...@imperial.ac.uk

personal web page: http://www.jfoadi.me.uk

 From: Jim Pflugrath <jim.pflugr...@rigaku.com>
Sent: Friday, 2 November 2012, 19:24
Subject: Re: [ccp4bb] Rpim and how its related to anomalous signal

In my opinion Rpim is not related directly to anomalous signal, so perhaps that 
is why there is some confusion. 

Also I think some folks confused Rpim with Rrim.  The latter is also called 
Rmeas.  But once again, these are not related directly to anomalous signal.  I 
do not find Rpim very useful for anything since when the data has high 
multiplicity Rpim gets very low, but as Michael Blum told me, "Precision does 
not trump accuracy."  I would personally rather have accurate data than highly 
redundant but inaccurate data.

I like to look at the deltaF / sigma(deltaF) value reported by SHELXC and other 
programs, where deltaF is ||F+| - |F-||.
This might also be called <d"/sig>.  There are other things to look at as well.


From: CCP4 bulletin board [CCP4BB@JISCMAIL.AC.UK] on behalf of Vijayakumar.B 
Sent: Tuesday, September 18, 2012 4:36 AM
Subject: [ccp4bb] Rpim and how its related to anomalous signal

Dear CCP4BB users,

I am very much interested to work in Anomalous scattering technique for 
macromolecular structure determination. I have already gone through some 
literature in which they have explained about a parameter called “Rpim”.  I am 
little bit confused about Rpim values.
Can anyone tell me about the use of this parameter in structure determination 
by anomalous techniques and how Rpim is related to anomalous signal?  Thanks in 

With kind regards

B. Vijayakumar

Reply via email to