Dear Patrick, Mark and colleagues,
Examining the data associated to PDB entry 2XGF using the PDBpeep
server, i.e. at
http://staraniso.globalphasing.org/cgi-bin/PDBpeep.cgi?ID=2xgf
shows a large region (marked up in dark blue) of missing measurements where
the analysis of the trends in the distribution of the local average I/sig(I)
would have led to expect statistically significant intensities. Such
systematic patterns of missing data are never good news when it comes to
maps.
This raises a couple of questions for you, Mark:
1. this is not the shape of a standard cusp, so would there have been
features of the crystal morphology that made some regions of reciprocal
space hard to reach in an experiment, apart from the usual cusp caused by
the 2-fold axis being too close to the rotation axis?
2. this pattern of missing data makes it rather surprising that the
completeness reported in the paper is about 99%, all the way out to the
outer resolution limit; the STARANISO logfile, accessible at
http://staraniso.globalphasing.org/PDB/2xgf.log
shows significantly lower values, even for the ellipsoidal completeness, of
about 0.7 at the highest resolution, for the deposited data.
Did you perhaps deposit only part of the data you collected for the
remote wavelength? For example, only one of several orientations that you
might have collected in order to try and fill the cusp?
This may seem a digression from the original question about negative
difference density, but clearly any systematically missing data can create
additional artefacts in maps and difference maps, besides those originating
from chemical or radiation damage related causes.
With best wishes,
Claus and Gerard.
--
On Mon, Feb 22, 2021 at 07:51:11PM +0100, Mark J van Raaij wrote:
> Hi Patrick,
> couldn't see any text in you email, just the subject and the picture, but
> 2XGF is our T4 long tail fibre tail tip structure from 2010.
> (2XGF stands for "2nd eXtraordinary Great Fibre", at least that's what I
> *modestly* imagine..).
> I think it's probably one or more of:
> - noise (was this contoured at 3sigma? or at a specific e/Å3 level? at the
> end of refinements 3sigma corresponds to less e/Å3 than at the start, or at
> least it should).
> - perhaps a small percentage of another metal than iron in the site (Co, Ni,
> Cu, Zn?), it's quite clear most of it was iron though from the X-ray
> fluorescence spectrum and the fact that the MAD phasing worked
> - perhaps some radiation damage
> - ripple effects around the heavy atom sites
> Resolution was 2.2Å and data was 99.9% complete, multiplicity 3.8, more
> statistics here: https://www.pnas.org/content/107/47/20287/tab-figures-data
> <https://www.pnas.org/content/107/47/20287/tab-figures-data> (paper
> https://www.pnas.org/content/107/47/20287).
> We didn't consider lowering the occupancy of the irons, or changing from Fe2+
> to Fe3+, because that would have meant less electrons, not more as suggested
> by the positive density.
> In any case, happy that this thread serves to learn something more from the
> comments by the experts.
> Mark
>
> Mark J van Raaij
> Dpto de Estructura de Macromoleculas
> Centro Nacional de Biotecnologia - CSIC
> calle Darwin 3
> E-28049 Madrid, Spain
> Section Editor Acta Crystallographica F
> https://journals.iucr.org/f/
>
>
> > On 22 Feb 2021, at 16:16, Patrick Needham <needh...@miamioh.edu> wrote:
> >
> >
> > To unsubscribe from the CCP4BB list, click the following link:
> > https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
> > <https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1><Screen
> > Shot 2021-02-22 at 10.08.36 AM.png>
>
>
> ########################################################################
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
>
> This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing
> list hosted by www.jiscmail.ac.uk, terms & conditions are available at
> https://www.jiscmail.ac.uk/policyandsecurity/
########################################################################
To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list
hosted by www.jiscmail.ac.uk, terms & conditions are available at
https://www.jiscmail.ac.uk/policyandsecurity/
