[Dock-fans] de novo Dock (beta) problems

Scott Brozell sbrozell at rci.rutgers.edu
Fri Jul 6 11:53:35 PDT 2018


Hi,

Thanks for the details.
It would still be useful to send me an example faulty Drugbank SDF
to improve our error reporting.

The grid file format was not changed from 6.8 to 6.9.

Others can comment on the other issues.

scott

On Fri, Jul 06, 2018 at 10:20:28AM +0100, Gyorgy Abrusan wrote:
> The crash was an error on my side, the SDF files that I downloaded from
> Drugbank apparently do not contain full 3D coordinates, and in consequence
> the converted mol2 files were also not correct. (I simply didn't check
> this, I assumed that the SDFs from Drugbank are fine. Sorry for this.)
> 
> The other issues do not seem to be my errors, so far I managed to complete
> a nova run only on a tiny fragment library, generated from a 100 compounds
> (from CACTVS smiles, and the generated compounds appear to be fine).
> Whenever I use bigger datasets for fragment library generation the
> libraries seem to be built without problems, but I get the following error
> message when I run the nova algorithm:
> 
> 
> -------------------------------------------------------------------
> ...
> Initializing Orienting Routines...
> Initializing Grid Score Routines...
>  Reading the energy grid from grid.nrg
>  Done reading the energy grid.
>  Reading the grid box quantifiers from grid.bmp
>  Done reading the grid box quantifiers.
> 
> 
> ##### Entering the main de novo build engine #####
> 
> Warning: atom type Mg in Fingerprint::return_atom_type not recognized
> ...
> Warning: atom type Mg in Fingerprint::return_atom_type not recognized
> 
> ##### Now orienting anchor #1
>   ##### Entering layer of growth #1
>   Root size at the beginning of this layer = 1
>     From Root[0], created 0 new fragments with an added layer
>       From all frags in root, kept a total of 0 new fragments with an added
> layer
>       After pruning by the RMSD heuristic, that number goes down to 0
>       After writing 0 of those to file, 0 are returned to root for the next
> iteration
> 
> Root is empty, continuing.
> 
> There are no more anchors. Exiting.
> 
> 
> 
> 0 Molecules Processed
> Total elapsed time:    9590 seconds
> -----------------------------------------------------------------------
> 
> The warnings about unrecognized atom types do not seem to be the cause of
> the error, fixing that results in the same error.
> 
> 
> I also pasted below the input file I use for the nova algorithm (without
> full paths), maybe that is the cause of the problem (I am using grid score
> as the primary score).  A third possibility is that using grids generated
> by dock6.8 is the cause, I have not checked this, so far in all of my runs
> I used my "old" grids, made with 6.8.
> 
> Best wishes,
> Gyorgy
> 
> 
> -----------------------------------------------------------------------------------
> conformer_search_type                                        nova
> dn_fraglib_scaffold_file
> .../fraglib/fraglib_scaffold.mol2
> dn_fraglib_linker_file
> .../fraglib/fraglib_linker.mol2
> dn_fraglib_sidechain_file
> .../fraglib/fraglib_sidechain.mol2
> dn_user_specified_anchor                                     no
> dn_use_torenv_table                                          yes
> dn_torenv_table
> .../fraglib/fraglib_torenv.dat
> dn_sampling_method                                           graph
> dn_graph_max_picks                                           30
> dn_graph_breadth                                             3
> dn_graph_depth                                               2
> dn_graph_temperature                                         100
> dn_pruning_conformer_score_cutoff                            100.0
> dn_pruning_conformer_score_scaling_factor                    2.0
> dn_pruning_clustering_cutoff                                 100.0
> dn_constraint_mol_wt                                         550
> dn_constraint_rot_bon                                        15
> dn_constraint_formal_charge                                  2.0
> dn_heur_unmatched_num                                        1
> dn_heur_matched_rmsd                                         2.0
> dn_unique_anchors                                            1
> dn_max_grow_layers                                           2
> dn_max_root_size                                             25
> dn_max_layer_size                                            25
> dn_max_current_aps                                           5
> dn_max_scaffolds_per_layer                                   1
> dn_write_checkpoints                                         yes
> dn_write_prune_dump                                          yes
> dn_write_orients                                             no
> dn_write_growth_trees                                        no
> dn_output_prefix                                             final
> use_internal_energy                                          yes
> internal_energy_rep_exp                                      12
> internal_energy_cutoff                                       100.0
> use_database_filter                                          no
> orient_ligand                                                yes
> automated_matching                                           yes
> receptor_site_file
> selected_spheres.sph
> max_orientations                                             1000
> critical_points                                              no
> chemical_matching                                            no
> use_ligand_spheres                                           no
> bump_filter                                                  no
> score_molecules                                              yes
> contact_score_primary                                        no
> contact_score_secondary                                      no
> grid_score_primary                                           yes
> grid_score_secondary                                         no
> grid_score_rep_rad_scale                                     1
> grid_score_vdw_scale                                         1
> grid_score_es_scale                                          1
> grid_score_grid_prefix                                       grid
> multigrid_score_secondary                                    no
> dock3.5_score_secondary                                      no
> continuous_score_secondary                                   no
> footprint_similarity_score_secondary                         no
> pharmacophore_score_secondary                                no
> descriptor_score_secondary                                   no
> gbsa_zou_score_secondary                                     no
> gbsa_hawkins_score_secondary                                 no
> SASA_score_secondary                                         no
> amber_score_secondary                                        no
> minimize_ligand                                              yes
> minimize_anchor                                              yes
> minimize_flexible_growth                                     yes
> use_advanced_simplex_parameters                              no
> simplex_max_cycles                                           1
> simplex_score_converge                                       0.1
> simplex_cycle_converge                                       1.0
> simplex_trans_step                                           1.0
> simplex_rot_step                                             0.1
> simplex_tors_step                                            10.0
> simplex_anchor_max_iterations                                500
> simplex_grow_max_iterations                                  500
> simplex_grow_tors_premin_iterations                          0
> simplex_random_seed                                          0
> simplex_restraint_min                                        no
> atom_model                                                   all
> vdw_defn_file
> .../dock6/parameters/vdw_AMBER_parm99.defn
> flex_defn_file
> .../dock6/parameters/flex.defn
> flex_drive_file
> .../dock6/parameters/flex_drive.tbl
> 
> 
> 
> On 5 July 2018 at 21:55, Scott Brozell <sbrozell at rci.rutgers.edu> wrote:
> > Thanks for the feedback.
> >
> > 1.  Can you send me a self contained set of input files to
> > reproduce the crash ?
> >
> > Someone will be in contact regarding the other issues.
> >
> > thanks,
> > scott
> >
> > On Thu, Jun 28, 2018 at 12:19:47PM +0100, Gyorgy Abrusan wrote:
> > > I have encountered the following problems with de novo dock
> > >
> > > 1. fragment library generation:
> > > a) if I convert the entries of drugbase from sdf to mol2 (with
> > openbabel),
> > > dock crashes, and is unable to build the libraries
> > > b) several atoms present in drugbase (and pdb ligands are not recognized
> > ,
> > > e.g. B, Ca, Co,Gd)
> > >
> > > 2. NOVA algorithm.
> > > It does seem to make a big difference how I create the fragment
> > libraries.
> > > 1) If convert canonical smiles from CACTVS to mol2 and use them for
> > > fraglib, the nova algorithm works and appear to generate valid ligands.
> > > 2) if I use different smiles to generate mol2, fragment library
> > generation
> > > is successful, but the nova algorithm does not seem to work with these
> > > libraries, and I get the error message that no anchors were found.
> > >
> > > (I used openbabel for all conversions)
> > >
> > > So I would like to ask for some advice about the recommended way of
> > > creating fragment libraries.


More information about the Dock-fans mailing list