| Home | News | Download | Features | Documentation | FAQ | Installation | Feedback |
Unpack the concoord distribution for your operating system in a directory of your choice ("tar xvzf concoord2.1_xxx.tar.gz"). If no concoord version is available for your system, let me know and we will try to provide you with a version. Go into the concoord directory and edit the CONCOORDRC according to the instructions inside. Close the editor and source the CONCOORDRC file. Now CONCOORD should be added to your environment and be ready for use.
The starting point of a CONCOORD simulation is a starting structure in PDB (.pdb) or GROMOS87 (.gro) format. Experimentally determined macromolecular structures can be obtained from the Protein Data Bank. Concoord has been primarily developed with proteins in mind, but also works with DNA/RNA, sugars and in principle any other molecule. Any CONCOORD simulation consists of two stages: structure interpretation and bond definitions (done by the program "dist") and structure generation based on these bounds ("disco"). For example, a simple CONCOORD simulation could therefore be performed using the following commands:
As of version 2.0, CONCOORD can be used to generate structures based on NMR-NOE data. A typical setup could look like:
First generate PSF file from PDB:
Thanks to Chris Spronk and Sander Nabuurs for the NMR module, including this conversion script, in CONCOORD.
disco does not generate any structures. What is the matter?
Can CONCOORD be used for structure regularisation?
Which set of VdW parameters do you recommend?
Which set of bonded parameters do you recommend?
I'd like to keep a certain part of my molecule fixed during structure
generation. How can I do that?
I get a segmentation fault as soon as I start a CONCOORD program. How come?
should I do an energy minimisation before running CONCOORD?
should I include hydrogens in my simulation?
should I use the -nb flag of dist?
should I use the -dssp flag of dist?
should I use the -bump option of disco?
how can I analyse CONCOORD results?
can I use CONCOORD for the simulation of point mutation effects?
I have a set of custom parameter (*.DAT) files. How can I use those without
having to change the global concoord library?
How can I cite CONCOORD?
The original CONCOORD method, with an introduction of the method and
first applications is described here:
(C) Bert de Groot, 1996-2010
Please note that the software is distributed with NO WARRANTY OF ANY KIND. The author is not responsible for any losses or damages suffered directly or indirectly from the use of the software. Use it at your own risk.
Please send your bug reports, comments and suggestions to:
Enjoy!
There can be several causes for this. Were there any dist/disco
warnings? If yes, please investigate them, they may provide hints to a
solution.
Are there many bad contacts in your input structure, or other weird
geometries? If so, it usually helps to start with a structure
regularisation or energy minimisation prior to running CONCOORD. Do
you use Engh/Huber bond lengths and angles? If yes, try with
CONCOORD's internal default parameters. Are there many "input
violations", as reported by disco? If yes, check what kind of
violations these are (disco -v). Do you have any atoms with zero
occupancy? (usually the second-last column in the starting PDB
file). If yes, was this intentional? Concoord interprets a zero
occupancy as an atom of which the position should remain fixed. This
usually makes convergence slower. If occupancies were unintentionally
set to zero, please change them into a non-zero value.
Also try increasing the number of constraints per atom (-m option of dist)
to e.g. 100.
Are you doing
NMR structure determination? Possibly not all NOE and geometric bounds can be
simultaneously fulfilled. Please increase the tolerances (bumps,
bonds, angles, planarities, NOE's) in MARGINS.DAT and increase the
overall level of acceptable violations (-viol command line switch of
disco), and try again. Note that the "-dyn" option of disco attempts to do
this automatically.
Although not primarily conceived for this purpose, it is in priciple
possible. Use the disco options -ref and -damp x (with x a value between 2 and 10)
to test this feature. Also, the -reg disco option can be used to regularise
concoord-generated structures.
The yamber2 and OPLS-AA parameters seem to work well for all-atom
simulations (including hydrogens). For other simulations, the OPLS-UA
parameters may be preferred.
Choosing Engh-Huber parameters results in slightly "better" structures
(in terms of PROCHECK and WHATCHECK scores), at additional
computational cost.
By setting the occupancies of the atoms to keep fixed to zero in the
starting PDB file.
Are you using an older concoord version? This should not happen with the 2.1
version. If the issue still occurs, try the following: Issue a "limit stacksize unlimited" to prevent
this. Repeat the procedure for "memoryuse" and "vmemoryuse". Bash users: replace
"limit" by "ulimit".
For good-quality or high-resolution structures it should not be
necessary. In case of slight irregularities in the starting structure,
energy minimisation may enhance convergence (see above).
including hydrogens will improve the quality of the generated
structures, but CONCOORD will run considerably slower. A
quick-and-dirty flexibility analysis, therefore, can generally be
carried out without hydrogens.
No, it is not generally recommended. Further development is currently
going on to make this a useful option in the future.
as of version 2.0, the dependence of CONCOORD results on DSSP
should be smaller than in previous versions, especially if hydrogens
are present in the starting PDB file (even if the -r flag of dist is
not used and the hydrogens are removed again by dist). The differences
have not been extensively tested yet, though, so please report your
experiences.
This is usually only required for NMR structure determination. But
it doesn't cost much extra in terms of CPU time, so it may also be a
good idea to switch it on for normal CONCOORD runs, as it
double-checks for bumps in this case.
CONCOORD results can be analysed in a variety of ways. Visual
inspection can be carried out using e.g.
Rasmol ,
VMD , Pymol, or gOpenmol
The GROMACS analysis tools can be
used for a convenient analysis of structures and trajectories,
including RMSD, RMSF, accessibility, secondary structure, gyration
radius and pricipal components (essential dynamics) analyses.
A: concoord generates based on a known structure, or "around" a known structure in
configurational space. Concoord won't, like e.g. MD, show a structural transition, or
drift due to e.g. a mutation. What can be seen usually is a shift if the average structure
due to a mutation (so between an WT and mutant ensemble). In a PCA analysis such
a shift can be recognised when the two trajectories are combined in one PCA and
differences in behaviour (averages, fluctuations) occur between the different trajectory
parts when projected onto single eigenvectors. These shifts along eigenvectors can
be small, but they may be significant, as concoord hardly suffers from a sampling problem
(when at least a few hundred structures are generated).
CONCOORD will first look in your local working directory for any
.DAT file. and use that if possible. If not present in your local
working directory, it will try the CONCOORD library (as defined in
$CONCOORDLIB). To prevent "dist" from copying files from the CONCOORD
library into your local working directory, use "dist.exe" instead of "dist".
If you use concoord for publications or presentations please cite the original
concoord publication:
B.L. de Groot, D.M.F. van Aalten, R.M. Scheek, A. Amadei, G. Vriend and
H.J.C. Berendsen; "Prediction of protein conformational freedom from distance
constraints", Proteins 29: 240-251 (1997)
[pdf]
"hidden" features
Documentation
B.L. de Groot, D.M.F. van Aalten, R.M. Scheek, A. Amadei, G. Vriend
and H.J.C. Berendsen;
"Prediction of protein conformational freedom from distance
constraints", Proteins 29: 240-251 (1997)
[pdf]
bgroot@gwdg.de.