Re: [ccp4bb] Is there a bulletin board (similar to ccp4bb) for small molecule crystallography

[Fred Vellieux]

Hello again,

I'll try ccp4 for Windows when I am back at home (where Windows "lives" 
in my case).


I have certainly tried shelxle. If I remember well what is provided is a 
.deb file. I used alien to convert that to a .rpm file, and installation 
failed because of issues I can't remember (the Linux flavour here is 
Alma Linux 9, not Debian).


Fred.

On 12/03/2024 10:56, David Waterman wrote:

Hi Fred,

CCP4 distributes the shelxl binary on Linux. I've not checked yet, but 
perhaps it is also part of CCP4 on Windows?


Cheers
-- David


On Tue, 12 Mar 2024 at 09:38, Fred Vellieux 
 wrote:


Hello and thanks for the reply.

My input file came from my checking SHELX manuals and SHELX
instructions on the web, and trying to modify them to suit my case.

I tried olex2 on a Windows computer (somehow olex2 did not run on
my Linux box, and only v1.3 was able to install on that Windows
machine and not the latest v1.5). However, olex2 (the installation
program) does not come with Shelx program executables, and I could
not locate the installers for the shelx software on Windows.

Hence I am running SHELXL in line command mode on my Linux box.

Thank you again,

Fred.

On 12/03/2024 10:12, David Waterman wrote:

Hi Fred,

I find Olex2 and shelxle are both convenient interfaces to SHELXL
refinement, that take care of some of the details of .ins file
format for you. However, maybe you are stuck in the starting
gate, depending on what is malformed in your input file. Where
did your input files come from?

Cheers
-- David


On Tue, 12 Mar 2024 at 09:01, Fred Vellieux
 wrote:

Hi folks,

I have a simple question: is there an electronic bulletin
board for
small-molecule crystallography? I have checked the list of
CCP projects
and there is no CCP-project for small molecule
crystallography in the list.

I am trying to run SHELXL, and it fails with the cryptic
message "** BAD
ATOM OR UNKNOWN INSTRUCTION **".

The alternative for me would be of course to use software
meant for
macromolecular cystallography (that I know) on small molecule
diffraction data. And using the small molecule coordinate files
transformed to a suitable format. I don't know if this is
feasible or
even advised. Probably not.

Thanks,

Fred.

-- 
MedChem, 1st F. Medicine, Charles University

BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1
<https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1>

This message was issued to members of
www.jiscmail.ac.uk/CCP4BB <http://www.jiscmail.ac.uk/CCP4BB>,
a mailing list hosted by www.jiscmail.ac.uk
<http://www.jiscmail.ac.uk>, terms & conditions are available
at https://www.jiscmail.ac.uk/policyandsecurity/

-- 
MedChem, 1st F. Medicine, Charles University

BIOCEV, Vestec, Czech Republic


--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

Re: [ccp4bb] Is there a bulletin board (similar to ccp4bb) for small molecule crystallography

[Fred Vellieux]

Hello and thanks for the reply.

My input file came from my checking SHELX manuals and SHELX instructions 
on the web, and trying to modify them to suit my case.


I tried olex2 on a Windows computer (somehow olex2 did not run on my 
Linux box, and only v1.3 was able to install on that Windows machine and 
not the latest v1.5). However, olex2 (the installation program) does not 
come with Shelx program executables, and I could not locate the 
installers for the shelx software on Windows.


Hence I am running SHELXL in line command mode on my Linux box.

Thank you again,

Fred.

On 12/03/2024 10:12, David Waterman wrote:

Hi Fred,

I find Olex2 and shelxle are both convenient interfaces to SHELXL 
refinement, that take care of some of the details of .ins file format 
for you. However, maybe you are stuck in the starting gate, depending 
on what is malformed in your input file. Where did your input files 
come from?


Cheers
-- David


On Tue, 12 Mar 2024 at 09:01, Fred Vellieux 
 wrote:


Hi folks,

I have a simple question: is there an electronic bulletin board for
small-molecule crystallography? I have checked the list of CCP
projects
and there is no CCP-project for small molecule crystallography in
the list.

I am trying to run SHELXL, and it fails with the cryptic message
"** BAD
ATOM OR UNKNOWN INSTRUCTION **".

The alternative for me would be of course to use software meant for
macromolecular cystallography (that I know) on small molecule
diffraction data. And using the small molecule coordinate files
transformed to a suitable format. I don't know if this is feasible or
even advised. Probably not.

Thanks,

Fred.

-- 
MedChem, 1st F. Medicine, Charles University

BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1
<https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1>

This message was issued to members of www.jiscmail.ac.uk/CCP4BB
<http://www.jiscmail.ac.uk/CCP4BB>, a mailing list hosted by
www.jiscmail.ac.uk <http://www.jiscmail.ac.uk>, terms & conditions
are available at https://www.jiscmail.ac.uk/policyandsecurity/


--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] Is there a bulletin board (similar to ccp4bb) for small molecule crystallography

[Fred Vellieux]

Hi folks,

I have a simple question: is there an electronic bulletin board for 
small-molecule crystallography? I have checked the list of CCP projects 
and there is no CCP-project for small molecule crystallography in the list.


I am trying to run SHELXL, and it fails with the cryptic message "** BAD 
ATOM OR UNKNOWN INSTRUCTION **".


The alternative for me would be of course to use software meant for 
macromolecular cystallography (that I know) on small molecule 
diffraction data. And using the small molecule coordinate files 
transformed to a suitable format. I don't know if this is feasible or 
even advised. Probably not.


Thanks,

Fred.

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

Re: [ccp4bb] Crystals with DNA

[Fred Vellieux]

Hello,

Overhanging "sticky" ends are mentioned frequently when it comes to 
obtaining infinite helices that are useful in crystallization. For 
example in 
https://home.ccr.cancer.gov/csb/nihxray/Tips-and-Tricks_Crystallization_Protein-DNA_updated.pdf 
.


Cheers,

Fred.

On 09/02/2024 10:59, careinaedgo...@yahoo.com wrote:

Thank you for this insight, Nicolas. It is very helpful.
Yes I have also had a soccer ball shaped crystal that does not 
diffract as well as, and more recently, many plate like crystals but 
they do not diffract either.
I do know I have both protein and DNA in my crystals but I do not 
know, as you say, exactly what is forming the crystal contacts.
Just to be clear, do you say overhangs are helpful? Surely overhangs 
won't promote an infinite helix? If one wants an infinite helix, would 
the DNA not have to be blunt ended?


Sent from Yahoo Mail on Android 



On Thu, Feb 8, 2024 at 4:59 PM, Nicolas Foos
 wrote:

Hello Careina,

In my hands, DNA protein complex crystals may be frustrating, 
because often we get good looking crystals which don't diffract at
all and are actually not easy to improve.

I remember obtaining a lot of crystal looking a bit like "STOP"
road sign (octogonal shape for one axis) which never diffracts.
(Often containing only DNA not well organized)

So long story short, In my hand (transciption factor bound with
homeodomain for example). I had good results with DNA sequence
which results in hoverhangs. The idea was to bet on a "infinit"
DNA helix which should help the packing.

I strongly encouraged you to rely on any other information  you
can have to be sure of what is the best minimal sequence (like
band shift assay). Also if you can purify the entire complex
before crystallization assay (I don't know your protocol, but
ideally, I would prepare the complex prot-DNA and put it on size
exclusion).

The point is, you don't know /a priori /what kind of crystal
packing you will have. It may be only due to protein protein
contact and not related to the DNA directly.

Also, I often get good results with crystal growing condition
containing MPD or PEG (makes me using PEG screen Familly as first
approach).

I invite you to read the Timothy Richmond teams Papers on 
nucleosome they spend some times improving the resolution on very
large complex. (Luger etal 1997).

There is many parameters, DNA sequence also change a bit the DNA
geometry (look for A-tract), You may want to introduce such
sequence to maybe improve the "rigidity".

Also if your DNA fragment are small, be careful with the
temperature. The annealing and the DNA duplex formation is
critical and you should be careful on your procedure.

I remember that small cation like Li, may help too.

HTH


Nicolas


On 08/02/2024 12:25, careinaedgo...@yahoo.com
 wrote:
 Hello all.

I am struggling to get defracting crystals with a protein DNA
complex. The crystals are plentiful but they do not diffract. I am
going back to the grind stone and relookong at my DNA sequence.
Is there any wisdom you could give me with regards to what works
best with DNA in crystals?
From my reading it seems if the length is a multiple of 7 (for B
DNA) and blunt ended, it will stretch over the length of the
crystal and improve crystalisability. But if you want crystals
that diffract better, you will need to play with length and even
making it only one base longer or shorter can make a difference,
even changing the morphology of the crystal? Longer is better than
shorter, and overhangs are good for improving diffraction?
Presumably because they stabilize contacts? It is expensive to
synthesize a while bunch of sequences so I need to be strategic in
my choice. Would appreciate any advice.
Thank you
Careina.



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1


-- 
Nicolas Foos PhD - ARISE fellow

https://orcid.org/-0003-2331-8399  


EMBL Grenoble, McCarthy Team

71 av. des Martyrs,
38000 Grenoble FRANCE

+33 4 57 42 84 67





To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1

Re: [ccp4bb] Solution to [ccp4bb] problem running ccp4mg: ugly and useless graphics

[Fred Vellieux]

Dear all,

In case anyone experiences problems with Alma Linux, NVIDIA graphics 
boards and the NVIDIA graphics driver, here is what worked:


somehow, the nouveau graphics driver seems to sneak back in, even if 
blacklisted. When nouveau is in use, the ugly graphics are obtained with 
ccp4mg.


There are several procedures mentioned on the web to install the NVIDIA 
driver.


This one worked, 
https://www.linuxcapable.com/how-to-install-nvidia-drivers-on-almalinux/ 
. All the nvidia or cuda packages installed by any alternative procedure 
must be removed first.


The other approaches mentioned on the web didn't work in my hands (and 
led to big problems).


The procedure mentioned in linuxcapable seems to set SecureBoot on, and 
then this HP Z4 computer doesn't boot any more. One has to enter the 
BIOS and have "Legacy support enable, Secure boot disable" set (and saved).


Just in case.

Fred.

On 17/01/2024 10:17, Frederic Vellieux wrote:

Dear all,

Perhaps a reader of the bb will have a solution for this problem. ccp4 
is version 8.0 and is up to date. All graphics programs (coot, but 
also other non-ccp4 software such as chimeraX, Pymol...) run fine on 
the Alma-Linux 9.3 system. There is a problem however with ccp4mg: the 
images appearing on the screen are... wrong and quite useless (see 
enclosed png screen capture). There is no error message at program 
startup.


I have disabled and blacklisted the nouveau graphics board driver and 
installed the NVIDIA driver. Made no difference.


So has anyone seen this behaviour before and has a solution to propose?

Thank you.

Regards,

Fred.




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/


--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

Re: [ccp4bb] problem running ccp4mg: ugly and useless graphics

[Fred Vellieux]

Dear Stuart,

I did test that (removing the .CCP4MG2 directory and its contents) and 
it gave the same ugly graphics.


There are problems with Alma Linux and the "Secure Boot" on this Hewlett 
Packard Z4 computer. This I have noticed. The secure boot option wants 
to load a signed graphics driver at boot time and I did not find a way 
to sign the NVIDIA driver (downloaded from the NVIDIA site as a .run 
file for installation) in a way that the system recognizes it. The 
recommended way of installing the NVIDIA driver on the Alma Linux pages 
doesn't allow to run later kernels when they become available (trying 
using dnf or yum and rebooting on a new kernel means a totally bricked 
computer).


So far the only option that seems to work is not to allow Secure Boot 
and not to allow Legacy settings. Otherwise the computer is bricked and 
requires a fresh installation of Linux. I don't know how many times I've 
had to re-install Alma Linux...


Jan Dohnalek mentioned that the ugly graphics could be due to a lack of 
memory problem. I am investigating that.


Cheers,

Fred.

On 1/17/24 11:47, Stuart McNicholas wrote:

Dear Fred,
  Many apologies for your difficulties. My first thought is that 
perhaps CCP4MG's lighting settings are completely messed up. This can 
be tested by removing completely the folder:


$HOME/.CCP4MG2

If this does not work, then further investigation will be required.

Also, in CCP4 we are working on a new graphics program: Moorhen. 
Currently most of our effort has been on model building aspects (it is 
based on Coot), but we are also working on figure preparation and it 
should over the next few months adopt most of the features of CCP4MG.


https://moorhen-coot.github.io/wiki/2023/11/03/Creating-Figures-with-Moorhen.html 
(tutorial)

https://moorhen.org/ (the program)

Best wishes,
Stuart


On Wed, 17 Jan 2024 at 09:17, Frederic Vellieux 
 wrote:


Dear all,

Perhaps a reader of the bb will have a solution for this problem.
ccp4
is version 8.0 and is up to date. All graphics programs (coot, but
also
other non-ccp4 software such as chimeraX, Pymol...) run fine on the
Alma-Linux 9.3 system. There is a problem however with ccp4mg: the
images appearing on the screen are... wrong and quite useless (see
enclosed png screen capture). There is no error message at program
startup.

I have disabled and blacklisted the nouveau graphics board driver and
installed the NVIDIA driver. Made no difference.

So has anyone seen this behaviour before and has a solution to
propose?

Thank you.

Regards,

Fred.




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/


--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] CentOS 7 end of life (july 2024)

[Fred Vellieux]

Hi,

Other people on this BB may run into the same problem.

CentOS 7 end of life is announced to happen in July 2024.

I have to migrate my Linux box to another Linux "flavour".

I've had a look at the possibilities:

- migrate to another RHEL (rpm-based) Linux, with "elevate-linux" and 
"leapp".
Here on this Linux box the problem I have is that the disk partition 
mounted as / uses btrfs. btrfs has been deprecated starting at versions 
8 (RHEL8, CentOS 8, Alma etc). This means first to copy all that is 
present on / somewhere, change the file system (for example to ext4) and 
restore everything.


- migrate to Debian, that supports btrfs. There are utilities, 
"debtakeover" and "debootstrap" that are supposed to install Debian 8.11 
(jessie).


Has anyone performed such a migration without data loss (files, 
pathways, configurations)? If so I'd like to know what was successful.


Thank you.

Fred.

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] Question about BIOVIA DiscoveryStudio 2021

[Fred Vellieux]

Folks, apologies for the non-CCP4 software question.

I have tried to contact the BIOVIA DiscoveryStudio support team, somehow 
my browser does not allow me to open their "contact form".


I am trying to visualize and perform an analysis on a protein:smaller 
molecule complex. The smaller molecule happens to be a long peptide. 
Whatever I do with the PDB (change ATOM cards to HETATM, change residue 
names to PEP, UNK, LGD, introduce MODEL and ENDMDL cards) 
DiscoveryStudio indicates the input PDB file doesn't contain any ligand.


Would any one in this community have an idea of how to specify that a 
part of a coordinate file is the ligand, so that DiscoveryStudio 2021 
recognises it as such and allows me to proceed?


Thanks,

Fred.

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

Re: [ccp4bb] how to get mol files from PDB and restraint CIF?

[Fred Vellieux]

Hello Frank,

We have to convert betwen file formats very frequently (usually several 
times daily) and:

1) we didn't need any restraints CIF file for that;
2) the tools we are using are
http://www.cheminfo.org/Chemistry/Cheminformatics/FormatConverter/index.html
https://datascience.unm.edu/tomcat/biocomp/convert

Sometimes the conversion doesn't take place and I have no idea why.

HTH,

Fred.

On 5/19/23 11:28, Frank von Delft wrote:
Hello - as in the subject line, does anybody know of, or have, code 
that will parse (1) a PDB (or mmCIF?) file with a ligand, and (2) the 
restraints CIF file used in refinement, and generate a .mol (or .sdf) 
file?


OpenBabel apparently does not.

I thought the PDB processing tools would, but my collaborator couldn't 
find them.


Any pointers welcome.  (To save us time.)

Thanks!
Frank



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/


--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

Re: [ccp4bb] ISO model with large groups of atoms in alternative conformations

[Fred Vellieux]

Hello Pavel,

You may want to have a look at PDB structure 6YCR: the macrocyclic 
peptide inhibitor is modelled as two conformations. The entire peptide.


HTH,

Fred.

On 8/24/22 00:02, Pavel Afonine wrote:


Dear community,

I’m looking for an example of a crystal structure where a large group 
of atoms (as large as a whole chain or even a domain) have more than 
one distinct conformation that would require modeling of such 
chain/domain as more than one individual copy, with each copy having 
partial occupancy. I’m not sure if that even exists but if someone can 
share an example that'd be very much appreciated!


Thanks!
Pavel




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1 




--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] SUMMARY: displaying residues (as a surface perhaps) for one component of a p-p-i

[Fred Vellieux]

Dear bb members,

My question from yesterday is repeated at the bottom of this email.

What worked is the suggestion by Jan Dohnalek (IBT and CMS, BIOCEV Vestec):

opening a structure file in CCP4MG with the option "interfaces" then 
selecting "residues". A simplified display of the proteins appears with 
the residues involved in the interface. This is followed by a PISA 
calculation from within CCP4MG. The surface of only one of the 
components in the p-p-i can be displayed, which is exactly what I was 
looking for.


Thanks for all the suggestions.

Regards,

Fred.

On 5/31/22 08:12, Fred Vellieux wrote:

Dear bb members,

I am quite certain that someone must have needed to do this already. I 
looked at publications but no details were given concerning how 
figures were prepared.


So here is the problem:

Given a protein protein interface (with the 3D structure of the 
complex available) I am looking for a method allowing me to identify 
and display the interface forming residues of one of the protein 
components. Plus a surface representation of that part, for the 3D 
structure.


Thanks in advance for any tips.

Regards,

Fred. Vellieux


--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] displaying residues (as a surface perhaps) for one component of a p-p-i

[Fred Vellieux]

Dear bb members,

I am quite certain that someone must have needed to do this already. I 
looked at publications but no details were given concerning how figures 
were prepared.


So here is the problem:

Given a protein protein interface (with the 3D structure of the complex 
available) I am looking for a method allowing me to identify and display 
the interface forming residues of one of the protein components. Plus a 
surface representation of that part, for the 3D structure.


Thanks in advance for any tips.

Regards,

Fred. Vellieux

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] Does anyone know of a "CHARMM bulletin board" ?

[Fred Vellieux]

Hi folks,

Sorry about the non-ccp4 question. I don't know where to ask.

I have to perform MD simulations but I am totally on my own here. The 
one program that appears to be free (in terms of license) seems to be 
CHARMM. GROMACS (testing with the examples found in the internet) did 
not work in my hands so I forget about that.


Charmm somehow looks similar to a program I used to work with, X-PLOR. 
However I haven't found a manual similar to the old XPLOR manual nor a 
web site that can generate the input files as was the case for X-PLOR. 
Neither did I find anywhere a flow chart allowing me to know which steps 
to be performed and in what order.


Hence: does anyone know of an X-PLOR to CHARMM dictionary ? Or a CHARMM 
bulletin board similar to CCP4BB ?


Just to give an example using the CHARMM setup.inp file for a protein 
that has an "ACE" cap at the N-terminus: fails. It complains about the 
ACE cap. Removing the cap solves that problem but then CHARMM fails 
because of "HIS". If CHARMM fails because it doesn't like any 
amino-acids then I don't know how I can use the program. But still I'd 
like to use it because I have to.


Thanks,

Fred.

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] cryptic error message trying to use LigPlot on an unknown ligand

[Fred Vellieux]

Hi folks,

I'm trying to run Ligplot on a PDB file that contains a residue with 
type UNL (Unknown Ligand). I hadn't been using LigPlot for perhaps 2 
years now (which means that I had to reinstall, with an expired license, 
and get familiar with it again). I get the following error message 
(rather cryptic):


Calling HBADD ...
Running HBADD
Het Group Dictionary: /components.cif
Temporary PDB file: /tmp/lig7141158588178475829/ligplus.pdb
Command:  /LigPlus/lib/exe_linux/hbadd 
/tmp/lig7141158588178475829/ligplus.pdb /components.cif -wkdir 
/tmp/lig7141158588178475829/
java.io.IOException: Cannot run program "/LigPlus/lib/exe_linux/hbadd": 
error=2, No such file or directory

Other event: state

Would anyone know what to make out of this message ? Otherwise is there 
another piece of software (called "app" nowadays) that could provide me 
with similar drawings ?


At some stage I ran PRODRG, introduced the cif file in the file 
components.cif used by LigPlot (LigPlus). I also replaced the coordinate 
files by those returned by the PRODRG run. Always with the same cryptic 
error message provided by the software (oops, app).


Thanks,

Fred.

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] coot & other graphics programs draw too many bonds

[Fred Vellieux]

Hello there,

After running autodock vina on certain small molecules, the graphics 
software I am using (e.g. Pymol, Coot) draws far too many bonds on the 
docked small molecule. See enclosed screen capture.


Is there any way to prevent this from happening? This isn't very 
satisfactory of you wish to produce figures for a presentation or for 
publication.


Ta,

Fred.

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

Re: [ccp4bb] Suggestions to improve resolution of protein crystals

[Fred Vellieux]

Hi,

In addition to all that's been suggested so far: when I was facing such 
problems I always tried the (commercially available) additive screens 
(additive kits, whatever they are called nowadays). This means setting 
up quite a few crystallisation experiments (you already have conditions 
in which crystals grow) and blindly going through the additives. Just in 
case a few of them improve the resolution of your crystals, or even 
provide another crystal form. I've had plenty of success this way. Of 
course the results obtained are not from sessions of hard thinking, what 
could be the cause of the problems with the current crystals etc but who 
cares? It may be that (for example) a small molecule inserts itself in 
the right place to give you the desired result.


Worth trying.

HTH,

Fred.

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] Help needed (input files)

[Fred Vellieux]

Hello,

I need to perform some MD calculations and then trajectories of some 
small molecules analyzed.


What I have is
1. protein
2. cofactor (FAD)
3. small molecule (either single O2 atom or single Chlorine atom)
4. crystallographic waters

The software I can access is either Gromacs (with yum install) or 
perhaps Desmond.


I have tried and tried to get this to run (for 6 months perhaps), to no 
avail. The input files located on the web do not work on the version of 
Gromacs provided by the yum install command. The Maestro license 
(required in order to get Desmond to run) is too expensive.


Is there a kind soul somewhere that would have suitable input files that 
would do all the above steps and who'd be willing to pass them on to me?


What needs to be done is:
add waters to the crystallographic waters in order to fill a box of 
suitable size

generate parameter and topology files for each component separately
merge these into global parameter and topology files for the system
run initial Energy Minimization
run M.D. simulations
analize the trajectories of the small molecules.

Thanks in advance.

Fred.

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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/

[ccp4bb] summary of replies, superimposition of 3D structures using the dsDNA part only

[Fred Vellieux]

Dear all,

I obtained the following suggestions to my query on the BB 
(superimposing two protein:dsDNA structures using the dsDNA structures 
alone for the superposition operation):


- Matthias Barone (fmp-berlin) suggested Chimera, with some instructions 
on "how to do it". He also suggested a program called moloc;


- Anat Bashan (Weizmann) suggested Coot, Calculate, LSQ Superimpose;

- Tim Gruene (univie) suggested Uppsala software factory's lsqman;

- Paul Emsley (MRC-LMB) mentioned that there was no "gesamt" for DNA 
(too bad) and further suggested lsq-improve;


- Jeremy (Tame ?, jt...@yokohama-cu.ac.jp) mentioned a program in c 
called cfit, which he can provide on request.


In the end I used Chimera using the instructions given by Matthias 
Barone and that worked like a charm. So to speak: the result obtained 
was what I wished to find out.


Thanks to all who replied. Superposition using the nucleid acid part of 
complexes can be very informative.


Have a nice day further,

Fred. Vellieux

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

[ccp4bb] superimposition of 3D structures with the DNA part only

[Fred. Vellieux]

Hi folks,

Some of you may have had to do this already. Either in the lab or more 
recently perhaps from home.


I have two structures that I wish to superpose (two protein:dsDNA 
complexes). Not using the protein part, but superposition through the 
dsDNA.


I'm not quite certain what is the "best" way of doing this.

Your suggestions will be appreciated, thanks.

Fred. Vellieux

--
MedChem, 1st F. Medicine, Charles University
BIOCEV, Vestec, Czech Republic



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] current location for the Superimposé web server ?

[Fred Vellieux]

Hello,

Follow up concerning my query to the bb (in case anyone is interested), 
reply received from Robert Preissner through Philip E. Bourne and Joel 
Sussman:


The Superimposé server was stopped in May 2018.

Fred. Vellieux


On 2/28/20 7:47 AM, Fred Vellieux wrote:


*Hello,

I am looking for the Superimposé web server (from Charité, Berlin).
*

*This web server must have moved from the place where it is supposed 
to reside, farnsworth.charite.de , not found.


Does anyone know where this web server resides now ?

TIA,

Fred.
*




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1





To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

[ccp4bb] current location for the Superimposé web server ?

[Fred Vellieux]

*Hello,

I am looking for the Superimposé web server (from Charité, Berlin).
*

*This web server must have moved from the place where it is supposed to 
reside, farnsworth.charite.de , not found.


Does anyone know where this web server resides now ?

TIA,

Fred.
*




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

[ccp4bb] Summary: MD program suitable to compute trajectories of a very small molecule in a protein

[Fred Vellieux]

Dear all,

Thank you for your help.

Here is a summary of the replies received:

---> Neli Fonseca, EBI, suggested the use of Docker containers,
"
https://hub.docker.com/search?q=gromacs=image

https://hub.docker.com/search?q=cp2k=image

https://hub.docker.com/search?q=nwchem=image "

---> Jeroen Mesters, Biochem Uni Lubeck and later Amit Singh pointed at 
the pkgs page for gromacs (Centos 7):


https://centos.pkgs.org/7/epel-x86_64/gromacs-2018.8-1.el7.x86_64.rpm.html

It then became clear why my attempt at "yum install gromacs" followed by 
"which gromacs" returned an error message (the command is "gmx");


---> Chris Roome, mpimf-heidelberg:
"If you're insterested, recently compiled the latest Gromacs 2020, seems 
to work fine.


I compiled GCC v5.5.0 first, and set the LD... and exe path (your path 
will differ of course) here's mine, with tcsh:


setenv LD_LIBRARY_PATH /software/gcc-5/lib64
set path = ( /software/gcc-5/bin $path )

As you say, even with this, cmake picks up the old, system installed 
gcc, so I had to specify on the cmake cmd:


cmake .. -DGMX_BUILD_OWN_FFTW=ON -DREGRESSIONTEST_DOWNLOAD=ON 
-DCMAKE_C_COMPILER=gcc -DCMAKE_CXX_COMPILER=c++ -DGMX_MPI=on 
-DCMAKE_INSTALL_PREFIX=/software/gromacs-2020 -DREGRESSIONTEST_DOWNLOAD


since cmake picks up the old cc even with the correct paths.

I installed the MPI version, so if you want that, you'll need to install 
the openmpi and openmpi-devel packages with yum. Then do a 'module load 
mpi' before the cmake."


--->  Abhik Mukhopadhyay suggested the use of NAMD, 
https://www.ks.uiuc.edu/Research/namd/


---> Eugene Osipov: you can try NAMD, Amber (cost-free CPU-only academic 
version) or Desmond - their installation is simple and they work 
reasonably well.


---> Jack Tanner (U. Missouri-Columbia): "You might want to review the 
literature on using MD to study diffusion of O2/CO in myoglobin." (I had 
started in fact)


[---> Lorenzo Briganti: the reply seems to have been lost along the way].

Hoping this will be useful to others, and thank you once again.

Fred. Vellieux, 1st Faculty of Medicine, Charles University in Prague, 
BIOCEV Vestec site




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

[ccp4bb] MD program suitable to compute trajectories of a very small molecule in a protein

[Fred Vellieux]

Dear all,

I need to run MD calculations in order to follow the trajectories of a 
very small molecule inside a protein. From previous calculations (not 
MD) I have starting positions for this small molecule that all seem in 
agreement with a possible path of motion inside the protein.


Now I need to access a MD program (without licensing costs).

I've had a look at the software list provided in Wikipedia (and tried to 
install the software, in succession, alas without success):


cp2k - present for CentOS6 (with yum install ?) but appears to have 
vanished for CentOS7;


gromacs requires a gcc version I don't have (even after having compiled 
and installed a suitable gcc version, cmake complains about the "old 
version" and stops);


NWChem is unhappy with the Python setup and doesn't compile.

I don't know how to solve all these OS version, library, unsuitable 
binary etc problems. In fact I don't know if any of these 3 software 
suites would be suitable for what I have in mind.


Hence would anyone know of a useful MD software suite that would be 
suitable for my purpose and comes with statically linked binaries 
suitable for Intel 64 (Linux) ? Just launch the executable and it runs, 
no questions asked...


Thank you,

Fred.



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

[ccp4bb] summary: how to define connectivity, bond types... in PDB files

[Fred Vellieux]

Dear all,

These are the replies received so far:

Daniel Rigden (U. of Liverpool) suggested to do the conversion using 
https://www.webqc.org/molecularformatsconverter.php . It didn't work for 
me, the small molecule may be too complex (error message when reading 
the .mol file);


Petr Kolenko (Czech Technical University in Prague) suggested to use the 
smiles small molecule definition format. molview.org provides a smiles 
string in the place where the URL appears when drawing the molecule and 
then converting it from 2D to 3D.


I located an online smiles translator 
(https://cactus.nci.nih.gov/translate/ ) and used it to generate a pdb 
file that appears to satisfy Pymol (see enclosed screen capture). I 
don't yet know if Pymol will still be happy after docking;


Stephane Rely (ENS Lyon) suggested the use of the prodrg server, and 
drawing the molecule with JME. I haven't tried this approach yet;


Finally, Massimo Degano (San Raffaele Scientific Institute in Milano) 
suggested to read in the pdb in Avogadro and writing it out, with CONECT 
cards written automatically. This didn't work on my system (Avogadro 
doesn't install due to Qt not being in found at the cmake stage, 
although it is present on my system)


Thank you for the prompt responses.

Fred.




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

[ccp4bb] Current "best" software for computing volumes of active sites

[Fred Vellieux]

Hi CCP4BBers,

With software constantly changing, new versions arriving to us and new 
software reaching the intended audience, I am looking for the most 
appropriate piece(s) of software to compute the volume of active site 
"cavities" in related 3D structures.


I have tried to use Voidoo however I do not know where to download an 
initial (and fairly comprehensive) cavity.lib file which I'd modify if 
need be. The links on the USF web site appear to be broken.


I am running Linux.

Thanks for the advice,

Fred. Vellieux



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] Backup of whole synchrotrons

[Fred Vellieux]

Hi,

We already have problems with the volume taken by our standard backups 
(they take too much space and we haven't been able to push the walls 
outwards in the Institute - I don't know why they keep telling us that 
our data should be in some clouds up in the sky). Hence I was wondering 
about space considerations: can the backups of the synchrotrons and 
X-FELs be miniature versions (obtained by clever dehydration methods)? 
If you need to access the backup then simply rehydrate and you'll get 
the full size backup appearing in the garden of your Institute... If 
your Institute doesn't have a garden of the proper size, then it's time 
to talk to the administrator. It should be fairly easy to convice 
her/him that the acquisition of a garden is really a must now.


F.

On 4/1/19 12:22 PM, Robbie Joosten wrote:

Hi Peter,

The copies are only indistinguishable after they have been produced. So there 
has to be good record keeping during production. It's as easy as hanging on to 
rich meta-data. There was another post today on what to store in mmCIF, I'm 
sure we can have another record in there to cover this.
You do touch the subject of FAIR data here, for reproducibility, do we need to 
keep the copy and spawn a new copy with the update? Or can we keep update the 
'original' copy with a well-defined downgrade path. Off course the meta-data 
for the original copy needs to be retained in such a case.
I hope it is obvious to everyone that we have to keep the copies in stasis, we 
cannot have them running around to change all the time. This is not just a 
methodological issue of being able to keep an experiment reproducible, but it 
is also an HR nightmare. It would require a lot of extra salaries. I mean, 
copies have rights and what will the unions think? If only we had thought of 
backing up crystallographers earlier, then we would have a copy of Margaret 
Thatcher to deal with the unions!  Instead, I guess today is a good day to 
invest in cryo-stasis technology.

Cheers,
Robbie

-Original Message-
From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of
Peter Keller
Sent: Monday, April 01, 2019 12:04
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] Backup of whole synchrotrons

Hi Robbie,

On 01/04/2019 07:23, Robbie Joosten wrote:

I don't think making this GDOR complient is that hard. It's all pretty
well defined what you store (everything), where you store it, and why.
There are some philosophical problems with allowing users to have
their data deleted. Assuming the copy is good enough to reproducing
the experiment. Deleting a copy would constitute murder.

You have correctly identified the underlying philosophical issue: it is a 
variant
of what is now known as the "Teletransportation paradox", see
.

  From the point of view of methods developers like you and me, there is an
additional issue: with insufficient raw data to work with, we are required to
create living experimenters as part of our development work.
For the most accurate results, these should be faithful copies of real
synchrotron visitors and beamline scientists, who in many cases are
personally known to us. How should we handle these copies when we need
to release new or updated methods? Since these copies need to be
indistinguishable from the originals, how can we tell whether we are
upgrading the copy or the original?

Regards,
Peter.

   This means that the

backups have to be stored in a rather libertarian "state" like Sealand
or Somalia.
Keeping that in mind, perrhaps this sort of backup should first be
implemented with the future African synchrotron.

Cheers,
Robbie

On 1 Apr 2019 07:46, "graeme.win...@diamond.ac.uk"
 wrote:

 While this may sound absurd, the principle of incremental backups
 can help out a great deal here. Like Apple’s Time Machine, all we
 need to do is store a copy of the things which have changed rather
 than the entire facility, which reduces the burden by at least a few
 orders of magnitude. Such efficiency savings will I am sure be of
 great interest in this project. Surely though we could save a copy
 of the experimental Eigenstate before the experiment too, offering
 the option of going back and having another go - every
 experimentalists dream!

 I do however take issue with your hypothesis that only the
 experimental equipment need be backed up - surely the experimenters
 also need to be archived, to allow the question “What were you
 thinking??” to be accurately answered when the reviewer’s questions
 come back. Unfortunately due to quantum entanglement issues this
 would probably require archiving the mind-state of dozens of people
 every time you hit “go” with the associated data protection issues -
 I for one would not like to fill in the GDPR section of that EU
 application :-)

 Anyhow, best of luck with your application,

 Graeme

 

Re: [ccp4bb] diagram of dsDNA

[Fred. Vellieux]

Wow, an email from the future !

[Sorry...]

 Subject:  [ccp4bb] diagram of dsDNA
 From: cuisheng2007 cuisheng2...@yahoo.com.cn
 Date: Mon, October 1, 2012 9:33 am
 To:   CCP4BB@JISCMAIL.AC.UK
 Dear all

 When I was playing around with Nucplot to generate diagram for
 dsDNA/Protein
 complex, the base pairs are displayed as letters. There must be a way of
 showing the bases as squares, a better illustration, but I cannot find a
 way
 of changing it in the nucplot.par. Does anyone know how to deal with it?



 Many thanks!

 Sheng



 Prof.Dr. CUI Sheng

 Institute of Pathogen Biology, CAMSPUMC.

 Building No.7, Bei Gong Da Ruan Jian Yuan,

 Beijing BDA, 100176,

 Beijing, China.

 Email: cuisheng2...@yahoo.com.cn

 Tel. +86 10 67828669

 Fax. +86 10 67855012








-- 
F.M.D. Vellieux (B.Sc., Ph.D., hdr)
IBS / ELMA
41 rue Jules Horowitz
38027 Grenoble Cedex 01
France
Tel: +33 438789605
Fax: +33 438785494

Re: [ccp4bb] NADP binding protein without Rossmann fold.

[Fred. Vellieux]

Hello,

You could look at this family of enzymes,
http://scop.mrc-lmb.cam.ac.uk/scop/data/scop.b.d.jc.b.b.html

[the sulfolactate dehydrogenase-like family]

No Rossman fold there.

HTH,

Fred.

 Dear all,

 I have biochemically characterized one enzyme that can dephosphorylate
 NADP+ / NADPH. Recently, I have also solved the crystal structure of the
 protein with bound NADP+. The important thing is that, the protein do not
 have the Rossmann fold or the dinucleotide binding fold. Can any one
 please
 cite or suggest any other example of such type of anomaly? Is there any
 example of non-classical Rossmann fold bearing proteins?

 Thanks,
 Sudipta.


 Sudipta Bhattacharyya.
 Senior Research Fellow,
 Department of Biotechnology,
 Indian Institute of Technology Kharagpur, India.



-- 
F.M.D. Vellieux (B.Sc., Ph.D., hdr)
IBS / ELMA
41 rue Jules Horowitz
38027 Grenoble Cedex 01
France
Tel: +33 438789605
Fax: +33 438785494

Re: [ccp4bb] Highest shell standards

[Fred. Vellieux]

On Fri, 23 Mar 2007, Edward Berry wrote:

 I believe fft does not by default do this- only if you use the fillin 
 keyword? One place where it might be important is in density 
 modification/ molecular averaging. Molecular averaging can be seen
 as a numerical solution of the MR equations, finding a density map
 which (A) obeys the NCS/intercrystal symmetry and (B) yields the 
 observed F's upon Fourier transformation.  Now if on each cycle
 you set the missing F's to zero, you are requiring it to have
 as part of (B) zero amplitude for the missing reflections, which is
 more restrictive and incorrect. If instead you allow the missing F's
 to float, calculating them on each cycle from the previous map
 using the fillin option, someone has shown (don't have the
 reference handy at the moment) that the F's tend toward the true F's
 (in the case that they weren't really missing but omitted as part
 of the test).
 
 Ed

You have phase scatter plots in Acta Cryst. D51, 575-589 (1995) that show
just that: the map inversion phases from NCS averaging tend toward the
true phases. Since F's are phased quantities and since phases are more
important than amplitudes, non random amplitudes plus non random phases
(both from map inversion of averaged maps) lead to better electron density
maps.

Fred.

-- 

 Fred. Vellieux, esq.

 =
 IBS J.-P. Ebel CEA CNRS UJF / LBM
 41 rue Jules Horowitz
 38027 Grenoble Cedex 01
 France
 Tel: (+33) (0) 438789605
 Fax: (+33) (0) 438785494
 =

Re: [ccp4bb] CNS v1.2 and unwanted introduction of OXT

[Fred. Vellieux]

On Mon, 5 Mar 2007, James Irving wrote:

 Hi Fred,
 
  From memory, this can occur due to long bond lengths in the model being 
 interpreted as chain breaks, try editing the generate.inp or 
 generate_easy.inp script and increase the value for break_cutoff:
 
 {* cutoff distance in Angstroms for identification of breaks *}
 {* the default of 2.5A should be reasonable for most cases. If the input
structure has bad geometry it may be necessary to increase this 
 distance *}
 {===} break_cutoff=2.5;
 
 You may also try switching off the automatic detection of mainchain breaks:
 
 * automatically detect mainchain breaks in proteins based on distance *}
 {* the peptide link at break points will be removed *}
 {+ choice: true false +}
 {===} auto_break=true;
 
 It is also worth checking that the offending residues have their full 
 complement of backbone atoms in appropriate places (not accidentally 
 zapped to far-flung regions of coordinate space).
 
 Cheers,
 James

Hi James,

I tried to increase the parameter value to 3.0 A. The resulting file gives
the same behaviour (and the newly introduced OXTs are within 0.05 A of the
N atom of the following residue) so I think the distances are reasonable.

So unfortunately I think your valuable suggestion does not solve the
problem. On the graphics with Coot the starting model looks fine in these
regions.

I need to keep the automatic mainchain break detection because there are
some gaps in each chain of the model (12 chains in all).

So if there are other suggestions to solve this br of a problem, they
are most welcome!

Fred.

-- 

 Fred. Vellieux, esq.

 =
 IBS J.-P. Ebel CEA CNRS UJF / LBM
 41 rue Jules Horowitz
 38027 Grenoble Cedex 01
 France
 Tel: (+33) (0) 438789605
 Fax: (+33) (0) 438785494
 =

Re: [ccp4bb] CNS v1.2 and unwanted introduction of OXT

[Fred. Vellieux]

On Mon, 5 Mar 2007, Fred. Vellieux wrote:

 Hi James,
 
 I tried to increase the parameter value to 3.0 A. The resulting file gives
 the same behaviour (and the newly introduced OXTs are within 0.05 A of the
 N atom of the following residue) so I think the distances are reasonable.
 
 So unfortunately I think your valuable suggestion does not solve the
 problem. On the graphics with Coot the starting model looks fine in these
 regions.
 
 I need to keep the automatic mainchain break detection because there are
 some gaps in each chain of the model (12 chains in all).
 
 So if there are other suggestions to solve this br of a problem, they
 are most welcome!
 
 Fred.

Hello again,

I believe that I found the cause of the problem and a solution to it:
not by looking at the molecule with Coot but by looking at the separate
PDB input files where the problems occur.

They all share the same features: the chain label used by CNS is kept
along within Coot. For the parts that are modelled de novo using Coot, the
output pdb file does not contain the CNS chain label. It is only at those
positions where there is a discontinuity for chain label present or absent
does CNS introduce an extra OXT atom. I have introduced the missing chain
labels where necessary, and I am testing this right now:

the file produced by the generate script now does not contain added OXT
atoms.

I thought I'd inform the bulletin boards of this annoying feature.

Fred.

-- 

 Fred. Vellieux, esq.

 =
 IBS J.-P. Ebel CEA CNRS UJF / LBM
 41 rue Jules Horowitz
 38027 Grenoble Cedex 01
 France
 Tel: (+33) (0) 438789605
 Fax: (+33) (0) 438785494
 =