Another reason to cut the data (temporarily!) is to avoid discussion later when you try to publish your model. You need to be able to defend your high resolution cut-off against the R-merge zealots. The paired refinement test to find a good resolution cut-off works well for that. It is based on model refinement and works best when your model is already quite good. Also, the test can only be unbiased if you haven't yet used your high-resolution data.
Whatever the outcome of your resolution cut-off choice, please deposit all recorded reflections!
Cheers,
Robbie
> -----Original Message-----
> From: CCP4 bulletin board [mailto:[log in to unmask]] On Behalf Of
> Alexandre Ourjoumtsev
> Sent: Friday, July 05, 2019 09:40
> To: [log in to unmask]
> Subject: Re: [ccp4bb] resolution
>
> Dear Graeme,
>
>
> Right, but you are talking about weights that reflect the data quality and say
> nothing about that of the starting model ; however refinement is a
> comparison of a model with data.
>
>
> The higher resolution of the data, the more sensitive they to model
> imperfections.
>
> Refinement targets are sums over reflections, and each refinement term is a
> function with multiple minima; the higher the resoluion, the more frequent
> these minima.
>
>
> If the starting model is too far from the answer, a presence of high-
> resolution data prevents the refinement from moving the model as far as
> necessary; it is trapped by multiple local minima of the crystallographic
> functions that include such high-resolution terms. Removing such terms
> removes or at least attenuate the intermediate local minima and improves
> the convergence. One does not care about the statistices but about
> convergence ("the model stops improving" further than with these data).
> Increaing the resolution step-by-step was the standard refinement strategy
> till the end of 90ths.
>
> Right, using ML-based targets introduced weights based on comparison of
> Fmodel with Fobs and allowed to do such attenuation in a "soft way". This
> was great and indeed replaced the "before-ML refinement strategy".
> However, such an artificial cut-off of highest-resolution data (temporary, at
> early refinement stages) can be useful in some situations even now and can
> improve convergence even with the modern tools. First cycles of a rigid-body
> refinement can be an example.
>
>
> Another reason for a (temporary) removing of higher-resolution data is a
> heavy (systematic) incompleteness of data in the higher-resolution shells.
>
>
>
> With best regards,
>
>
> Sacha
>
>
> ----- Le 5 Juil 19, à 8:05, [log in to unmask]
> <[log in to unmask]> a écrit :
>
>
> Pavel,
>
> Please correct if wrong, but I thought most refinement programs
> used the weights e.g. sig(I/F) with I/F so would not really have a hard cut off
> anyway? You’re just making the stats worse but the model should stay ~ the
> same (unless you have outliers in there)
>
> Clearly there will be a point where the model stops improving, which
> is the “true” limit…
>
> Cheers Graeme
>
>
>
> On 5 Jul 2019, at 06:49, Pavel Afonine
> <[log in to unmask]<mailto:[log in to unmask]>> wrote:
>
> Hi Sam Tang,
>
> Sorry for a naive question. Is there any circumstances where one
> may wish to refine to a lower resolution? For example if one has a dataset
> processed to 2 A, is there any good reasons for he/she to refine to only, say
> 2.5 A?
>
> yes, certainly. For example, when information content in the data
> can justify it.. Randy Read can comment on this more! Also instead of a hard
> cutoff using a smooth weight based attenuation may be even better. AFAIK,
> no refinement program can do this smartly currently.
> Pavel
>
> ________________________________
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-
> bin/webadmin?SUBED1=CCP4BB&A=1
>
>
> --
> 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
>
>
> ####################################################
> ####################
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-
> bin/webadmin?SUBED1=CCP4BB&A=1
>
>
>
> ________________________________
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB&A=1
|