Analysis in water refinement produced bad pdb

HADDOCK
1

Hi all,

I am having trouble figuring out why my analysis folder of water refinement produced weird pdbs, whereas the pdbs in the water folder are normal. I use local installation of haddock2.2. For example the best scoring structure 51w.pdb has the following coordinates in the beginning:

ATOM 1 CB SER 29 34.909 9.442 -22.209 1.00 10.00 A
ATOM 2 OG SER 29 36.028 10.056 -22.823 1.00 10.00 A
ATOM 3 HG SER 29 36.816 9.526 -22.648 1.00 10.00 A
ATOM 4 C SER 29 32.646 9.918 -21.234 1.00 10.00 A
ATOM 5 O SER 29 32.959 9.567 -20.095 1.00 10.00 A
ATOM 6 N SER 29 34.152 11.751 -21.916 1.00 10.00 A
ATOM 7 HT1 SER 29 34.357 11.838 -20.899 1.00 10.00 A
ATOM 8 HT2 SER 29 33.412 12.429 -22.169 1.00 10.00 A

the corresponding pdb fit_1.pdb has the following coordinates:
ATOM 1 CB SER 29 24.194 33.504-185.172 1.00 0.00 A
ATOM 2 OG SER 29 22.191 32.633-187.295 1.00 10.00 A
ATOM 3 HG SER 29 21.726 32.088-188.890 1.00 10.00 A
ATOM 4 C SER 29 29.658 37.145-175.596 1.00 10.00 A
ATOM 5 O SER 29 29.702 37.547-177.179 1.00 10.00 A
ATOM 6 N SER 29 22.966 33.259-185.233 1.00 10.00 A
ATOM 7 HT1 SER 29 22.213 33.881-186.570 1.00 10.00 A
ATOM 8 HT2 SER 29 21.627 32.387-185.663 1.00 10.00 A

Can it be that the 8th column has coordinates that have too many number of digits? I am not sure what is the problem. Suggestions much appreciated!

Yuan

2

The PDBs in the analysis dir have been fitted onto the average structure, which is why the coordinates are different.
Did you try viewing them and superimposing them for example in PyMol?

3

Hi Prof. Bonvin,

Yes I tried chimera and pymol and they all look like fragmented atoms scattered everywhere including the ave.pdb… There is occasional tyrosine side-chain ring that is intact but floating by itself. That is why I suspected that something is off with the coordinates after analysis. The ##w.pdb files were read normally by chimera and pymol. The ave.pdb looks similar to the fit_1.pdb:

ATOM 1 CB SER 29 24.194 33.504-185.172 0.00 0.00 A
ATOM 2 OG SER 29 22.191 32.633-187.295 0.00 10.00 A
ATOM 3 HG SER 29 21.726 32.088-188.890 0.00 10.00 A
ATOM 4 C SER 29 29.658 37.145-175.596 1.00 10.00 A
ATOM 5 O SER 29 29.702 37.547-177.179 0.00 10.00 A
ATOM 6 N SER 29 22.966 33.259-185.233 1.00 10.00 A
ATOM 7 HT1 SER 29 22.213 33.881-186.570 0.00 10.00 A
ATOM 8 HT2 SER 29 21.627 32.387-185.663 0.00 10.00 A

The rmsave.disp file also have some very small rms calculated with values around 5E-04.

Thanks,
Yuan

4

Strange… My guess is that you have some “exploded” models in the water directory.
Simply look at the water models and not the analysis ones.
Note that ...fit_1.pdb in the analysis directory corresponds to the first ranked model in the water dir.

5

Hi Prof. Bonvin,

Indeed very strange! As you suggested, I was able to look at all twenty structures in water directory by chimera and pymol, without spotting any exploded structures. Anything I missed to investigate?

Thanks,
Yuan

6

For this one you can simply focus on the water models.
Was this run performed on the web server or on a local version?

Without looking more into your data, we can’t tell much what is going on here.