Dear Jacob,
DIALS and MOSFLM each output both of the intensities you describe (although DIALS does 3D profile fitting whereas MOSFLM does 2D profile fitting). The first, Isum, are the summation integrated intensities, the second, Iprf, are the profile-fitted intensities. The scaling program AIMLESS (by default) combines the two values by calculating a weighted mean that varies smoothly from Iprf for the weak reflections to Isum for the strong reflections (it turns out that summation integrated intensities are often better than the profile-fitted intensities for strong data). See the AIMLESS documentation or Phil Evan’s scaling papers for more details:
http://www.ccp4.ac.uk/html/aimless.html#intensities
http://journals.iucr.org/d/issues/2013/07/00/ba5190/index.html#SEC2
Cheers,
Richard
On 10 Feb 2016, at 17:09, Keller, Jacob <[log in to unmask]<mailto:[log in to unmask]>> wrote:
Let me clarify: I was not asking whether there is such a thing as profile fitting, 3D or otherwise. I thought, however, that profiles used for making a mask through which to integrate the actual counts, and those numbers were the intensities. Not true? Are the derived mathematical profile estimates/fits themselves output as the measurements?
A difference practically between the two is whether fine phi slicing would reduce error by allowing multiple measurements of the Gaussian for each spot. In terms of fitting, it would be much better to sample the Gaussian ten times at different phi than once (back to the observation:parameter ratio question--a ratio of ten is better than one). Thus with fitting, one could consider each spot, if measured 10 times, to have a sort of multiplicity of its own.
JPK
-----Original Message-----
From: [log in to unmask]<mailto:[log in to unmask]> [mailto:[log in to unmask]]
Sent: Wednesday, February 10, 2016 9:38 AM
To: Keller, Jacob; [log in to unmask]<mailto:[log in to unmask]>
Subject: RE: Spot Integration Vs Fitting
Dear Jacob
There is a strong tradition of profile fitting in integration - for 3D profile fitting I would feel that this is the best place to start:
http://scripts.iucr.org/cgi-bin/paper?S0021889888007903
(Kabsch, 1988)
This was about processing data from a multiwire detector which behaved a lot like the pixel array detectors of today re: fine slicing & big (ish) pixels
The assumption of a Gaussian form is however a dangerous one, as we are certainly seeing "fine structure" in spots these days.
I am certain you will get many other references pointed your way, going back to e.g. Diamond 1969 on the profile fitting for point detector data
Best wishes Graeme
________________________________________
From: CCP4 bulletin board [[log in to unmask]] on behalf of Keller, Jacob [[log in to unmask]]
Sent: 10 February 2016 14:24
To: ccp4bb
Subject: [ccp4bb] Spot Integration Vs Fitting
Dear Crystallographers,
As I understand it, all integration software uses measured intensities rather than fits thereof. Wouldn't it be better in the case of (very) fine phi slicing to start using 3D gaussian fits to the spots, perhaps even with outlier rejection? I would think a fit to, say, 10 samples of a Gaussian would be more precise than summing the intensities.
Perhaps a lot of datasets have non-Gaussian distribution?
JPK
*******************************************
Jacob Pearson Keller, PhD
Looger Lab/HHMI Janelia Research Campus
19700 Helix Dr, Ashburn, VA 20147
email: [log in to unmask]
*******************************************
--
This e-mail and any attachments may contain confidential, copyright and or privileged material, and are for the use of the intended addressee only. If you are not the intended addressee or an authorised recipient of the addressee please notify us of receipt by returning the e-mail and do not use, copy, retain, distribute or disclose the information in or attached to the e-mail.
Any opinions expressed within this e-mail are those of the individual and not necessarily of Diamond Light Source Ltd.
Diamond Light Source Ltd. cannot guarantee that this e-mail or any attachments are free from viruses and we cannot accept liability for any damage which you may sustain as a result of software viruses which may be transmitted in or with the message.
Diamond Light Source Limited (company no. 4375679). Registered in England and Wales with its registered office at Diamond House, Harwell Science and Innovation Campus, Didcot, Oxfordshire, OX11 0DE, United Kingdom
|