· Delarue web servers|
· Nomad Flow-chart
· Normal Mode calculation
· Examples (Movies)
· Submit a job (from PDB file)
· Split trajectory (for MR)
· Generate decoys (MixMod)
· Elastic Energy (Perturb. Anal.)
· Overlap coefficients
· Submit a job
· Include Profit
· X-Ray refinement
· Standard refinement
· Screening Mol. Repl. Solns.
· EM refinement
· Get Structure Factors
· Submit a job (no NCS)
· Submit job with NCS
· Docking refinement
· Submit a job
· Force field methods
· Energy minimization
· Gromacs NMA
· Job queue status
Performs refinement of a model against X-ray data
along Normal Mode directions of low frequency. You must provide:
The model can contain either only CA atoms or all atoms. In the
latter case you can choose either to calculate exact normal modes
for the large system (slower) or use the RTB method that introduces
an efficient block approximation. RTB is significantly faster, and the
low frequency modes should be very similar between the two methods.
The code for RTB is from Y.-H. Sanejouand (firstname.lastname@example.org)
and the code for all-atoms NMA is from E. Lindahl (email@example.com).
The engine of refinement is Conjugate-Gradient refinement.
- The resolution limits of the data.
- The number of modes to use (increasing frequency order).
- The distance weight for the spring constant.
- The cutoff radius for pairs of atoms to be considered.
Input data formats
- The job title is just for your own identification, but note that
it will show up in the public job queue (but your results will not be public).
- The coordinate file should be in PDB format, with only a single
structure (no multiple models). Atoms marked with alternate residue
flags will be removed. By default the refinement only uses C-alpha atoms,
and the refined output will only consist of C-alpha coordinates.
- The symmetry data is a text file that describes the unit cell and symmetry operators.
On the first line there should be six numbers, representing unit cell dimensions (a,b,c)
in Ångströms and the cell angles (alpha,beta,gamma) in degrees.
The format is '(6f8.3)'.
This is followed by symmetry operators, using the AMoRe program format.
Each operator is followed by a '*', and the list is terminated with an 'end' record.
If there is more than one line of symm. op. to be read, the program should be able to
handle this without problem.
However, we have sometimes observed troubleshooting; if it happens, make sure each line is terminated with a '*'
and fill in the rest of the line with blanks until it reaches 80 characters.
A couple of examples (you should NOT just copy these):
- The reflection data is a free-format text file with one
reflection per line. Columns are separated by at least one space, and
there are 5 columns per line. The first
three are integers corresponding to the h,k,l indicies. Column 4 is
the amplitude of the reflection and column 5 the error (sigma). Column five
is currently not used, but the code expects a floating-point value to be there.
Reflections outside the low/high limits specified by you will be discarded.
An example: Fobs_example.data.
Parameters and options
- The refinement is carried out in reciprocal normal mode space.
Low-frequency modes offer the advantage to allow for collective and large-
In the order of 10 modes should work well for normal proteins, but you
can experiment with higher values, although the execution will be slower.
- All interactions are weighted by exp(-(r/r0)^2), where r0 is a
distance-weight parameter. Higher values lead to a stiffer system, while
lower values decouple local and global motions. For an example, consider
all-atom modes in a protein. A low value of r0 will create larger
hydrogen motions relative to the heavy atoms, even for low frequency
eigenmodes. A value of 3.0 Angstrom works well for CA-only
models, but you might prefer a larger value (5.0-10.0 Angstrom) for
See Hinsen article in References.
- A cutoff is used in the mode calculation. In the Tirion model
(Elastic Network Model) only those pairs of atoms that are closer than
the cutoff are linked by a spring of universal length. Ideally this should
be choosen so that the weighting causes the interactions to be negligible
outside the cutoff, but in practice a cutoff around 10 Angstrom works
fine in almost all cases. In general you might be able to use a smaller
cutoff value for all-atom calculations than CA-only calculation.
See Tirion article in References.
- The fastest option is to only use alpha carbon atoms, which also
means you can work on much larger systems. If you want to use all-atom
models in the refinement there are two options: either we can determine
the full all-atom eigenmodes, or we can use the RTB algorithm (rotation-
translation blocks) to calculate approximate all-atom eigenmodes faster.
See Tama and Sanejouand article in References.
- You should remove any hydrogen atoms from the coordinate files, since
their fast motions usually deteriorate the performance of the refinement.
- You can pre-scan each normal mode amplitude in a one-dimensional
search prior to Conjugate-gradient refinement, especially if you think that
the model has an rmsd larger than 4-5 Angstrom compared to the solution.
This will take a while with all atoms but is recommended in a first run using CAs
If you feel the maximum amplitude allowed for the movements is not high enough,
send me an e-mail and I will change it for you.
See also Claude et al. article in References.