Dear Charles,
I am doing water refinement (using Xplor 2.46) for a small metalloprotein,
after structure calculation using metal restraints in CYANA.
In my protein, I have the metal cluster which topology I "freeze" in the
wrefine.py script by fixing metals and binding atoms.
The problem I am experiencing is that for certain models I get
extremely large total energies even though all the models are very similar.
For example:
Average values for potential terms:
type name Energy(dev) RMSD(dev) viols(dev)
Num
PotList TOTAL 18046.88(84910.86) 7.602( 0.746) 1620.3(
84.4) 4203
PotList noe 15.03( 8.72) 0.031( 0.007) 0.1( 0.4)
504
XplorPot ANGL 138.18(402.79) 0.745( 0.800) 9.4( 4.4)
1312
XplorPot BOND 326.98(1387.25) 0.011( 0.021) 0.3( 1.6)
732
XplorPot CDIH 0.77( 0.84) 0.393( 0.181) 0.1( 0.4)
68
XplorPot DIHE 752.18( 69.71) 42.858( 0.498) 1130.5( 10.7)
1196
XplorPot ELEC -1638.22( 72.43)
XplorPot IMPR 686.14(2970.74) 1.573( 3.149) 16.6( 3.7)
391
XplorPot VDW 17765.82(80021.66) 463.3( 80.3)
sort fit comparison
Filename: energy RMSD RMSD
wrefine_1.sa -1008.96 0.810 2.661
wrefine_8.sa -999.73 0.771 1.073
wrefine_14.sa -984.83 1.001 1.593
wrefine_2.sa -966.03 0.532 2.096
wrefine_16.sa -963.98 0.880 2.172
wrefine_3.sa -958.11 0.898 1.743
wrefine_19.sa -952.42 0.776 2.677
wrefine_17.sa -950.52 1.058 2.290
wrefine_5.sa -948.20 0.797 1.495
wrefine_6.sa -945.71 0.808 1.273
wrefine_0.sa -934.16 0.831 2.549
wrefine_13.sa -933.36 0.787 1.377
wrefine_7.sa -928.58 0.664 1.210
wrefine_4.sa -922.42 1.033 1.782
wrefine_11.sa -920.08 0.775 0.919
wrefine_18.sa -909.48 0.659 2.448
wrefine_12.sa -887.80 0.700 1.031
wrefine_15.sa -881.41 1.145 1.830
wrefine_9.sa -859.76 0.645 0.870
wrefine_10.sa 378793.10 0.724 0.908
During calculation I get also a message:
Powell::step: irregular exit: Line search abandoned: gradient may be
incorrect
*-- POWELL ------ step= 1 --- stepsize= 0.01000 --- energy evals=
5 -*
| E(poten)=389918.3302198 grad= 30044.6033617 ANGL=
1619.5106059 |
| BOND= 6321.6578947 CDIH= 1.0494587 DIHE=
1028.5803754 |
| ELEC=-11596.8057242 HARM= 0.0000000 IMPR=
132.7689804 |
| VDW=392361.9271258 noe= 49.6415031
|
*------------------------------------------------------------------------------*
*--- Dynamics ---- step= 0 ---- time= 0 ---- delta_t=
0.002 --*
| E(kin)+E(poten)= 389918.330 E(kin)= 0.000 temperature=
0.000 |
| E(poten)=389918.3302198 grad= 30044.6025470 ANGL=
1619.5106059 |
| BOND= 6321.6578947 CDIH= 1.0494587 DIHE=
1028.5803754 |
| ELEC=-11596.8057242 HARM= 0.0000000 IMPR=
132.7689804 |
| VDW=392361.9271258 noe= 49.6415031
|
*------------------------------------------------------------------------------*
InternalDynamics::step: large timestep detected. Halving.
Do you know what could be a problem and how could I solve it?
Kind regards,
Jelena
_______________________________________________
Xplor-nih mailing list
[email protected]
https://dcb.cit.nih.gov/mailman/listinfo/xplor-nih
