DOCK 6.12 This is a release of DOCK6 with a new sampling method for hierarchical traversal through precomputed ligand conformations, as is done in DOCK 3.7. We have also updated the Chemgrid (DOCK3.5) scoring function to agree with DOCK3.7. Both of these changes are presented in reference [1]. DOCK_DN sampling has been rewritten to allow more molecules to be constructed in a similar amount of time, with no duplicates within a given run. Additional features are the covalent docking algorithm, called attach-and-grow, and the new scoring function GIST Score both of which are descibed in [2]. NEW IN DOCK 6.12 Implementation of Hierarchical database (HDB) search New features include the Hierarchical database (HDB) search [1], an alternative method to anchor-and-grow, to search the ligand degrees of freedom. This is the method used in DOCK 3. This includes a db2 reader and a search routine for hierarchical traversal through precomputed ligand conformations, stored in the db2 files, and it enables large-scale docking. We provide example scripts for building ligands as db2 files (and mol2 files) here: $DOCK6PATH/template_pipeline/hdb_lig_gen/ And we are also providing example scripts for large-scale docking here: $DOCK6PATH/template_pipeline/VS_submit_scripts/ Implementation of Attach-and-grow We implemented a covalent ligand sampling method called attach-and-grow [2]. In this method the covalent residue is defined by spheres (for cysteine C_alpha, C_beta, and S_gamma) and dummy atoms on the ligand are aligned to these spheres. The molecule is then grown using the anchor-and-grow machinery (the grow_periphery function). For minimization we use only torsion premin; so, only the ligand internal degrees of freedom are explored. Rework of pruning scheme for DOCK_DN The pruning scheme in DOCK_DN has been modified from a 'serial' approach, where molecules were only HRMSD pruned within the same attachment, to a 'parallel' approach, where the entire ensemble is pruned at multiple stages. This allows for many more molecules to be constructed in a similar time, ie higher efficiency, with no duplicates in the final output. Overhaul on DOCK_DN output The output of the DOCK_DN has been overhauled to be cleaner and more informative. It keeps track of the number of attachments DOCK_DN is working on at the root sampling level. Also it keeps track of the number of molecules being written out to the environment. Further, pruned molecules will no longer be output to a single file. They will be outputted based on where DOCK6 is pruning those molecules. Dockerization of DOCK6 for better installation experience While DOCK6 has been relatively easy to compile in Linux-based systems, there has been many issues with compatibility with other operating systems. Hence, an installation protocol that involves the Dockerization of DOCK6 and its tools has been developed. This new tool will help streamline the installation in almost any computational environment as long as the user has the Docker engine installed. Update of DOCK3.5 score DOCK3.5 score, also called chemgrid score, was updated to match that in DOCK 3.7 [1]. And it was integrated into descriptor score. Implementation of GIST score GIST score is a way to calculate receptor desolvation [2]. It was integrated into descriptor score to allow it to be used with other Scoring functions including grid score and chemgrid score. Miscellaneous Final min was added back in for mirroring DOCK 3.7 behavior. A score cut off was added to prevent poorly scoring molecules from being written. A parameter was added to write per-atom desolvation values (values read in from mol2 or db2 files), so DOCK3.5 (chemgrid) score may be used to rescore these poses. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ BUG FIXES None. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ INPUT PARAMETER CHANGES AND THEIR DEFAULT VALUES Example: parameter_name [default_value] (options): Added: For HDB: conformer_search_type HDB num_per_search 1 (an integer) skip_broken yes (yes no) hdb_db2_input_file ligand.db2.gz (this is a db2 file or a sdi.txt file) hdb_db2_search_score_threshold 10.0 For GIST: gist_score_primary yes gist_score_att_exp 6 gist_score_rep_exp 12 gist_score_gist_scale -1 gist_score_gist_type blurry_displace [trilinear] (trilinear displace blurry_displace) gist_score_grid_file gist-EswPlusEww_ref2.dx gist_score_blurry_gist_div 2.0 For Attach-and-grow: conformer_search_type covalent covalent_bondlength 1.6:0.1:2.0 # covalent_bondlength format options: # 'blstart:blstep:blstop' ; 'blstart:blstop' ; or bl # number of values = 3 covalent_bondlength2 -1.0 # covalent_bondlength2 format options: # 'blstart2:blstep2:blstop2' ; 'blstart2:blstop2' ; or bl2 # number of values = 1 # just on bond length2 is used -1 # covalent_bondlength2 < 0.0: input bond length will be used for D1--A1. covalent_angle -1.0 # covalent_angle format options: # 'avstart:avstep:avstop' ; 'avstart:avstop' ; or av # number of values = 1 # just on angle_val is used -1 # angle_val < 0.0: input angle length will be used for D2--D1--A1 .) covalent_dihedral_step 10.0 For DOCK_DN: dn_remove_duplicates yes (yes no) dn_max_duplicates_per_mol 0 dn_write_pruned_duplicates no (yes no) dn_advanced_pruning yes (yes no) ~with advanced pruning on~ dn_prune_initial_sample yes (yes no) dn_sample_torsions yes (yes no) dn_prune_individual_torsions yes (yes no) dn_prune_combined_torsions yes (yes no) dn_random_root_selection no (yes no) For descriptor_score: descriptor_use_gist no (yes no) descriptor_use_dock3.5 no (yes no) --- Descriptor Score Parameters: GIST Score --- descriptor_gist_score_att_exp 6 descriptor_gist_score_rep_exp 12 descriptor_gist_score_gist_scale 1 descriptor_gist_score_gist_type trilinear (trilinear displace blurry_displace) descriptor_gist_score_grid_file gist-EswPlusEww_ref2_cap_blury_1p8.dx descriptor_gist_score_Hydrogen_grid_file gist-EswPlusEww_ref2_cap_blury_1p0.dx --- Descriptor Score Parameters: DOCK3.5 Score --- descriptor_dock3.5_vdw_score [yes] (yes no): descriptor_dock3.5_grd_prefix [chem5] (): descriptor_dock3.5_electrostatic_score [yes] (yes no): descriptor_dock3.5_ligand_desolvation_score [no] (no total volume): descriptor_dock3.5_receptor_desolvation_score [no] (yes no): descriptor_dock3.5_write_atomic_energy_contrib [no] (yes no): descriptor_dock3.5_score_vdw_scale [1] (): descriptor_dock3.5_score_es_scale [1] (): --- Descriptor Score Weights --- descriptor_weight_gist_score -1 descriptor_weight_dock3.5_score [1] (): Miscellaneous automated_matching_iteration yes (yes no) simplex_final_min no (yes no) write_mol_solvation no (yes no) score_threshold 100.0 (real number) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ OUTPUT COMPONENTS For HDB: 1. In both the dockout file and the mol2 header the following components are written: HDB_Conf_Number: 11 HDB_Internal_energy1: 4.530000 HDB_Internal_energy2: 0.000000 This is tne conf_number from the db2 file and the internal energy values also read in from the db2 file. for GIST Score 1. In both the dockout file and the mol2 header the following component is written: GIST_Score: -0.494120 For Database Filter: 1. ${output}.rejected.mol2 is the mol2 file containing scored molecules that got rejected by dbfilter. For overhauled output DOCK_DN: 1. output.completed.denovo_build.mol2 2. ${output}.anchor_{#}.filtered_comp_layer_{##}.mol2 3. ${output}.anchor_{#}.prune_root_layer_{##}.mol2 4. ${output}.anchor_{#}.prune_rmsd_mw_layer_{##}.mol2 5. ${output}.anchor_{#}.cand_root_ign_layer_{##}.mol2 6. ${output}.anchor_{#}.root_layer_{##}.mol2 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ KNOWN BUGS 1. When selecting both internal energy and internal energy score, the parameter "internal_energy_rep_exp" appears twice in the question tree, and it will be queried during user input when DOCK is invoked from the command-line even if the value is already set. 2. Scoring functions are only able to handle receptors with less than 2000 atoms. It's possible to use more, but that may introduce undefined behavior in the current implementation. If users wish to use more, they should contact the developers. 3. In the flex.defn file, there is a flag minimize which is disabled. (The flag has no impact on the calculation.) All torsions are Minimized. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ RELEASE DATE September 2024 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ REFERENCES 1. Balius, T.E.; Tan, Y.S.; Chakrabarti, M.; DOCK 6: Incorporating hierarchical traversal through precomputed ligand conformations to enable large-scale docking. J. Comput. Chem. 2024, 45(1), 47-63. https://doi.org/10.1002/jcc.27218 2. Tan, Y.S.; Chakrabarti, M.; Stein, R.M.; Prentis, L.E.; Rizzo, R.C.; Kurtzman, T.; Fischer, M.; Balius, T.E.; Development of receptor desolvation scoring and covalent sampling in DOCK 6: Methods evaluated on a RAS test set. Submitted. 2024. https://chemrxiv.org/engage/chemrxiv/article-details/66db73a512ff75c3a19f4370