Distance geometry simulated annealing


Here a structure is calculated using experimentally measured interproton distance estimates, hydrogen bonds and coupling-constant-derived dihedral angle restraints. This protocol uses ab simulated annealing starting from embedded substructures using distance geometry calculations (based on the experimental data). The structure calculation is performed with the CNS task file dg_sa.inp.

      cns_solve < dg_sa.inp > dg_sa.out [25 minutes]

In this protocol the extended coordinate template is used as a starting point for generation of an embedded structure. This embedded structure is generated using distance geometry calculations such that the coordinates satisfy the known geometric and experimental distrance restraints. The resulting coordinates need to be further regularized with a simulated annealing protocol. In this case five trial structures were generated. An alternative approach would have been to only generated accepted structures, in which case the script would have run until five accepted structures had been generated.

The embedded structures are written out first, prior to simulated annealing refinement. Here the names are trx_dg_sub_embed_[number].pdb. In this case, the backbone trace of the coordinates shows the overall fold of the macromolecule:

Five structures calculated by distance geometry embedding.

After generating the embedded structures, they are further refined using simulated annealing. In this case the resulting structures have a greater variation than those generated by ab initio simulated annealing:

Five structures after simulated annealing refinement.

A summary of the structure calculation is written at the top of each output PDB file:

REMARK The macromolecule has 118 residues
REMARK Trial structure 1 of 5 structures
REMARK Molecular dynamics scheme : cartesian; cartesian; minimize
REMARK High temperature dynamics :
REMARK      temp: 2000 steps: 1000 time(ps): 3
REMARK 1st cooling stage         :
REMARK      temp: 2000->0 steps: 1000 time(ps): 5 temp step: 25
REMARK 2nd cooling stage not used:
REMARK a total 2000 steps of minimization
REMARK VDW scale factors 20 20 0.01 0.003 0.003; 3E-03->4; 1
REMARK 2617 NOEs in 2 class(es) with scale factors of 50; 50; 50
REMARK      averaging function(s): sum, hbnd: sum
REMARK 0 3-bond j-couplings in 0 class(es) with
REMARK      scale factor(s) of NA
REMARK 0 1-bond j-couplings in 0 class(es) with
REMARK      scale factor(s) of NA
REMARK 0 carbon chemical shifts in 0 class(es) with
REMARK      scale factor(s) of NA
REMARK 0 proton chemical shifts in 0 class(es) with
REMARK      scale factor(s) of NA
REMARK 0 diffusion anisotropy restraints in 0 class(es) with
REMARK      scale factor(s) of NA
REMARK 0 susceptability anisotropy restraints in 0 class(es) with
REMARK      scale factor(s) of NA
REMARK 325 dihedral restraints with scale factors of 5; 200; 400
REMARK 0 planarity restraints with a scale factor of NA
REMARK NCS restraints not used.
REMARK ===============================================================
REMARK               bond, angles, improp, vdw(<1.6),  dihed
REMARK violations :     2       8       1         2      41
REMARK RMSD       : 0.0034   0.565   0.420            23.671
REMARK ===============================================================
REMARK                noe,   cdih,   coup,   oneb, carb-a, carb-b,
REMARK violations :     4       1       0       0       0   -----
REMARK RMSD       :  0.030   0.513   0.000   0.000   0.000   0.000
REMARK 0.2/2 viol.:     9       5       0
REMARK ===============================================================
REMARK               dani,   sani
REMARK violations :     0       0
REMARK RMSD       :  0.000   0.000
REMARK .2/.1 viol.:     0       0
REMARK ===============================================================
REMARK Protons       violations, rmsd
REMARK all    :           0       0.000
REMARK class 1:           0       0.000
REMARK class 2:           0       0.000
REMARK class 3:           0       0.000
REMARK class 4:           0       0.000
REMARK ===============================================================
REMARK overall = 428.304
REMARK bon     = 21.554
REMARK ang     = 162.677
REMARK imp     = 25.2858
REMARK vdw     = 96.3987
REMARK harm    = 0
REMARK noe     = 111.966
REMARK coup    = 0
REMARK oneb    = 0
REMARK carb    = 0
REMARK prot    = 0
REMARK dani    = 0
REMARK sani    = 0
REMARK cdih    = 10.4229
REMARK ncs     = 0
REMARK ===============================================================
REMARK DATE: 8-Jul-99  16:54:04       created by user: paul
REMARK VERSION:0.5
ATOM      1  CA  MET     1      10.622  -0.138   7.804  1.00  0.00      A
ATOM      2  HA  MET     1      10.509  -1.030   8.403  1.00  0.00      A

The information about violations can be used manually to select acceptable structures. Instead the CNS task file accept.inp can be used to identify acceptable structures and to calculate an average conformation (see previous tutorial)


Script to run this tutorial

Back to tutorials   Previous section