[gmx-users] GMX-compatible DNA coordinates

[Mauricio Esguerra Neira]

Alex,


You can use X3DNA to produce any nucleic acid geometry you’d like and
gromacs pdb2gmx will understand it.

For example, to create a single-strand of 16 DNA bases you only need a file
with the corresponding rigid-body base-step parameters such as the
ssdna.par file shown below:

  16 # bases
   0 # ***local step parameters***
#      Shift     Slide     Rise      Tilt      Roll      Twist
T      0.000     0.000     0.000     0.000     0.000     0.000
A      0.280    -0.664     3.201     2.499    -2.895    32.486
C     -0.314    -0.644     2.640     8.643     2.631    28.956
C      0.424    -0.849     3.240     1.756     1.690    33.400
A      1.519    -0.396     3.077     4.184     6.152    39.140
A      0.412    -0.670     2.797     6.577    11.309    26.608
C     -0.183    -2.185     2.859     6.133     5.126    17.125
A      0.423    -1.477     6.195   -18.545   -26.691    36.074
C      1.216    -0.879     3.063     5.417     8.466    35.858
A     -9.111    -4.794    -5.433    74.829    43.216   -56.814
C      3.440     0.344     3.176     3.036    12.621    42.046
A     -9.111    -4.794    -5.433    74.829    43.216   -56.814
C      3.440     0.344     3.176     3.036    12.621    42.046
A      1.478     0.681     3.056    13.088     0.208    44.226
A     -0.358    -0.777     3.494    -2.803    -4.534    33.162
A     -0.264    -0.272     2.833     9.637     1.457    30.508



With such a file at hand all you have to do is issue the following two
commands to 3DNA.

    x3dna_utils cp_std -d BDNA

Which copies a standard BDNA backbone scaffold in the folder you’re at, and
then:

    rebuild -atomic ssdna.par ssdna.pdb

This will generate the file ssdna.pdb which you can find at:

http://wikisend.com/download/614970/ssdna.pdb

It works with:

    pdb2gmx -f ssdna.pdb -o start.pdb -p topol.top -i posre.itp -ff
charmm27 -water tip3p

In principle the online version of 3DNA, w3dna, would also work for
generating nucleic acid models of any desired sequence. But sadly the X3DNA
version working in the backend of the webserver is awfully outdated and
uses the old convention of simply A, G, T, C, instead of DA, DG, DT, DC, to
name the nucleotide residues.


Hope this helps,


Mauricio



+++++++++++++++++++++++++++++++++++++++++++++++
 Mauricio Esguerra Neira, Ph.D.
 Post-Doctoral Fellow
 Hugo Gutiérrez de Terán P.I.
 Department of Cell and Molecular Biology
 Uppsala University
 Phone Number: (+46)073-678-5334
 mesguerra.org
 mauricio.esguerra at icm.uu.se
+++++++++++++++++++++++++++++++++++++++++++++++