Small molecule docking with xtal waters

HADDOCK
1

I’m getting the following error for all it1 structures (see below)

my ‘protein1’ contains MG+2 and the xtallographic waters.

Any idea what is wrong?
Is there a protocol somewhere for ligand docking keeping
all or some of the xtallographic waters from the original
structure? Will those waters be distinguishable from the
added waters during water refinement in it1/water?

thanks, E

[…]POWELL: STEP number limit. Normal termination
POWELL: Current coordinates set to last minimum
--------------- cycle= 4 --------------------------------------------------
| Etotal =37061.567 grad(E)=159.952 E(BOND)=843.065 E(ANGL)=931.171 |
| E(DIHE)=1155.713 E(IMPR)=35861.318 E(VDW )=-1146.305 E(ELEC)=-866.604 |

E(NOE )=283.209

CALLING flex_segment.cns FOR RIGID BODY SA
NO FLEXIBLE SEGMENTS for all molecule at this stage
NO FULLY FLEXIBLE SEGMENTS for molecule 1
NO FULLY FLEXIBLE SEGMENTS for molecule 2
FLEXIBILITY STATISTIC FOR MOLECULE 1 : Nres=331 Nrigid=331 Nflex=0
FLEXIBILITY STATISTIC FOR MOLECULE 2 : Nres=1 Nrigid=1 Nflex=0
NUMTREES set to 23
DTOREST2: total number of groups 152
DTOREST2: total number of joints 0
DTOREST2: total number of closed loops 0
DTOREST2: total number of distinct bodies 152
DTOREST2: total number of chains 0
DTOREST2: maximum number of chains (per distinct body) 0
DTOREST2: maximum unbranched chain length 0
DTOREST2: adj. max. number of chains (per distinct body) 2
DTOREST2: adj. maximum unbranched chain length 2
-------------------------- Torsion Topology -----------------------------------
Please send a bug report to brunger@stanford.edu
Please increase the “maximum number of distinct bodies” (torsion_maxtree) and re-run.
%TORSION:TOPOLOGY error encountered: Fatal Topology Error
(CNS is in mode: SET ABORT=NORMal END)

ABORT mode will terminate program execution.

Program will stop immediately.
============================================================
Maximum dynamic memory allocation: 1071496712 bytes
Maximum dynamic memory overhead: 4896 bytes
Program started at: 13:02:46 on 26-Aug-2016
Program stopped at: 13:02:56 on 26-Aug-2016

2

The maximum number of tree is defined in HADDOCK in the numtree.cns script. It does count the number of chains and water molecules and add to that an additional buffer of 20. How many waters to you keep? The error message report 152 groups, which seems quite a lot.

I would remove everything not around the binding site.

3

ok that seems to work, except that the waters are discarded in the water refined structures,
probably together with all the other waters.

E

Alexandre Bonvin wrote:

4

You have to turn on the option to write additional pdbs containing the waters. Can be found in the analysis parameters menu of the guru interface:

Check the box:

After the final solvent refinement, write additional PDB files including solvent