Molecular docking methods for assessing complementarity of combinatorial libraries to biotargets

Abstract

A high-throughput molecular docking facility is presented for screening combinatorial libraries to identity binding ligands and ultimately pharmaceutical compounds. The facility employs a pre-coking conformational search to generate multiple solution conformations of a ligand. The molecular docking facility includes: generating a binding site image of the target molecule, the binding site image to atoms in at least one solution conformation of the multiple solution conformations of the ligand to obtain at least one ligand postion relative to the target molecule in a ligand-target molecule complex formation; and optimizing the at least one ligand position while allowing translation, orientation and rotatable bonds of the ligand to vary, and while holding the target molecule fixed. Docking results are clustered using as a metric the rms deviation between the core of two docked molecules. A library is rated as to complementarity to the target molecule according to the relative number of ligands in the top cluster.

What is claimed is:  
     
         1 . A method of docking a ligand to a target molecule comprising: 
 performing a pre-docking conformational search to generate multiple solution conformations of the ligand;    generating a binding site image of the target molecule, said binding site image comprising multiple hot spots;    matching hot spots of the binding site image to atoms in at least one solution conformation of the multiple solution conformations of the ligand to obtain at least one ligand position relative to the target molecule in a ligand-target molecule complex formation; and    optimizing the at least one ligand position while allowing translation, orientation and rotatable bonds of the ligand to vary, and while holding the target-molecule fixed.    
     
     
         2 . The method of  claim 1 , wherein said performing the pre-docking conformational search comprises creating a database of the multiple solution conformations and storing said three-dimensional database for subsequent use by said matching.  
     
     
         3 . The method of  claim 2 , wherein said database of the multiple solution conformations comprises a conformational database of a combinatorial library.  
     
     
         4 . The method of  claim 1 , wherein said performing the pre-docking conformational search comprises: 
 randomly generating a plurality of uniformly distributed conformations of the ligand;    minimizing a strain of each potentially active conformation;    using the strain and one or more three-dimensional descriptors for each conformation to rank the potentially active conformations; and    clustering the conformations and retaining a desired number of top clusters of conformations, said retained number of top clusters of conformations comprising said multiple solution conformations of the ligand.    
     
     
         5 . The method of  claim 4 , wherein one or more three-dimensional descriptors are selected from the group consisting of a polar solvent accessible surface area, an apolar solvent accessible surface area, number of internal interactions, radius of gyration, and combinations thereof.  
     
     
         6 . The method of  claim 4 , wherein said one or more three-dimensional descriptors are a combination of the polar solvent accessible surface area and the apolar solvent accessible surface area.  
     
     
         7 . The method of  claim 1 , wherein said generating the binding site image includes at least one of creating a list of apolar hot spots identifying points in the binding site that are favorable for an apolar atom to bind, and generating a list of polar hot spots identifying points in the binding site that are favorable for a hydrogen bond donor or acceptor to bind.  
     
     
         8 . The method of  claim 7 , wherein said generating the binding site image further comprises: 
 placing a grid around the binding site of the target molecule;    determining a hot spot search volume using said grid;    determining hot spots using a grid-like search of the hot spot search volume; and    for each type of hot spot, clustering the hot spots and retaining a desired number of clusters of hot spots with best scores, said desired number of clusters comprising said multiple hot spots to be employed by said matching.    
     
     
         9 . The method of  claim 1 , wherein said matching comprises: 
 matching atoms of the at least one solution conformation to appropriate hot spots of the target molecule by positioning the at least one solution conformation as a rigid body into the binding site image;    defining a match, said match determining a unique rigid body transformation; and    using the unique rigid body transformation to place the at least one solution conformation of the ligand into the binding site of the target molecule.    
     
     
         10 . The method of  claim 9 , wherein said determining the unique rigid body transformation comprises determining the unique rigid body transformation that minimizes:  
       
         
           
             
               
                 I 
                  
                 
                   ( 
                   
                     R 
                     , 
                     T 
                   
                   ) 
                 
               
               = 
               
                 
                   ∑ 
                   
                     j 
                     = 
                     1 
                   
                   3 
                 
                  
                 
                   
                      
                     
                       
                         H 
                         j 
                       
                       - 
                       
                         RA 
                         j 
                       
                       - 
                       T 
                     
                      
                   
                   2 
                 
               
             
           
           
           
               
           
         
         where: 
 H j =a j th  hot spot of the target molecule;  
 A j =a j th  atom of the at least one solution conformation;  
 R=a 3×3 rotation matrix; and  
 T=a translation vector.  
 
       
     
     
         11 . The method of  claim 1 , wherein said optimizing comprises optimizing multiple target molecule-ligand complex formations, said optimizing comprising: 
 eliminating each ligand position having a predetermined percentage of ligand atoms with a steric clash;    ranking remaining ligand positions using an atom pairwise score with a desired atom score cutoff;    after ranking, clustering the ligand positions and selecting a top number n of ligand positions; and    optimizing each ligand position of the n positions, allowing the translation, rotation and rotatable bonds of the ligand to vary.    
     
     
         12 . The method of  claim 11 , wherein said optimizing comprises optimizing each ligand position of the n positions using a BFGS optimization algorithm with a simple atom pairwise score, allowing the translation, rotation and rotatable bonds of the ligand to vary.  
     
     
         13 . A system for docking a ligand to a target molecule comprising: 
 means for performing a pre-docking conformational search to generate multiple solution conformations of the ligand;    means for generating a binding site image of the target molecule, said binding site image comprising multiple hot spots;    means for matching hot spots of the binding site image to atoms in at least one solution conformation of the multiple solution conformations of the ligand to obtain at least one ligand position relative to the target molecule; and    means for optimizing the at least one ligand position while allowing translation, orientation and rotatable bonds of the ligand to vary, and while holding the target molecule fixed.    
     
     
         14 . The system of  claim 13 , wherein said means for performing the pre-docking conformational search comprises means for creating a database of the multiple solution conformations and for storing said three-dimensional database for subsequent use by said matching.  
     
     
         15 . The system of  claim 14 , wherein said database of the multiple solution conformations comprises a conformational database of a combinatorial library.  
     
     
         16 . The system of  claim 13 , wherein said means for performing the pre-docking conformational search comprises: 
 means for randomly generating a plurality of uniformly distributed conformations of the ligand;    means for minimizing a strain of each conformation of the plurality of uniformly distributed conformations;    means for using the strain and a solvent accessible surface area of each conformation to rank the conformations; and    means for clustering the conformations and retaining a desired number of top clusters of conformations, said retained number of top clusters of conformations comprising said multiple solution conformations of the ligand.    
     
     
         17 . The system of  claim 13 , wherein said means for generating the binding site image includes at least one of means for creating a list of apolar hot spots identifying points in the binding site that are favorable for an apolar atom to bind, and means for generating a list of polar hot spots identifying points in the binding site that are favorable for a hydrogen bond donor or acceptor to bind.  
     
     
         18 . The system of  claim 13 , wherein said performing the pre-docking conformational search comprises: 
 means for randomly generating a plurality of uniformly distributed conformations of the ligand;    means for using a three-dimensional descriptor for each conformation to distinguish potentially active conformations from inactive conformations, and retaining potentially active conformations;    means for minimizing a strain of each potentially active conformation;    means for using the strain and a solvent accessible surface area of each potentially active conformation to rank the potentially active conformations; and    means for clustering the conformations and retaining a desired number of top clusters of conformations, said retained number of top clusters of conformations comprising said multiple solution conformations of the ligand.    
     
     
         19 . The system of  claim 18 , wherein said three-dimensional descriptor is selected from the group consisting of a polar solvent accessible surface area, an apolar solvent accessible surface area, number of internal interactions and radius of gyration.  
     
     
         20 . The system of  claim 17 , wherein said means for generating the binding site image further comprises: 
 means for placing a grid around the binding site of the target molecule;    means for determining a hot spot search volume using said grid;    means for determining hot spots using a grid-like search of the hot spot search volume; and    for each type of hot spot, means for clustering the hot spots and for retaining a desired number of clusters of hot spots with best scores, said desired number of clusters comprising said multiple hot spots to be employed by said matching.    
     
     
         21 . The system of  claim 13 , wherein said means for matching comprises: 
 means for matching atoms of the at least one solution conformation to appropriate hot spots of the target molecule by positioning the at least one solution conformation as a rigid body into the binding site image;    means for defining a match, said match determining a unique rigid body transformation; and    means for using the unique rigid body transformation to place the at least one solution conformation of the ligand into the binding site of the target molecule.    
     
     
         22 . The system of  claim 21 , wherein said determining the unique rigid body transformation comprises determining the unique rigid body transformation that minimizes:  
       
         
           
             
               
                 I 
                  
                 
                   ( 
                   
                     R 
                     , 
                     T 
                   
                   ) 
                 
               
               = 
               
                 
                   ∑ 
                   
                     j 
                     = 
                     1 
                   
                   3 
                 
                  
                 
                   
                      
                     
                       
                         H 
                         j 
                       
                       - 
                       
                         RA 
                         j 
                       
                       - 
                       T 
                     
                      
                   
                   2 
                 
               
             
           
           
           
               
           
         
         where: 
 H j =a j th  hot spot of the target molecule;  
 A j =a j th  atom of the at least one solution conformation;  
 R=a 3×3 rotation matrix; and  
 T=a translation vector.  
 
       
     
     
         23 . The system of  claim 13 , wherein said means for optimizing comprises means for optimizing multiple target molecule-ligand complex formations, said means for optimizing comprising: 
 means for eliminating each ligand position having a predetermined percentage of ligand atoms with a steric clash;    means for ranking remaining ligand positions using an atom pairwise score with a desired atom score cutoff;    after ranking, means for clustering the ligand positions and selecting a top number n of ligand positions; and    means for optimizing each ligand position of the n positions, allowing the translation, rotation and rotatable bonds of the ligand to vary.    
     
     
         24 . The system of  claim 23 , wherein said means for optimizing comprises means for optimizing each ligand position of the n positions using a BFGS optimization algorithm with a simple atom pairwise score, allowing the translation, rotation and rotatable bonds of the ligand to vary.  
     
     
         25 . At least one program storage device readable by a machine, tangibly embodying at least one program of instructions executable by the machine to perform a method of docking a ligand to a target molecule, comprising: 
 performing a pre-docking conformational search to generate multiple solution conformations of the ligand;    generating a binding site image of the target molecule, said binding site image comprising multiple hot spots;    matching hot spots of the binding site image to atoms in at least one solution conformation of the multiple solution conformations of the ligand to obtain at least one ligand position relative to the target molecule; and    optimizing the at least one ligand position while allowing translation, orientation and rotatable bonds of the ligand to vary, and while holding the target molecule fixed.    
     
     
         26 . The at least one program storage device of  claim 25 , wherein said performing the pre-docking conformational search comprises creating a database of the multiple solution conformations and storing said three-dimensional database for subsequent use by said matching.  
     
     
         27 . The at least one program storage device of  claim 26 , wherein said database of the multiple solution conformations comprises a conformational database of a combinatorial library.  
     
     
         28 . The at least one program storage device of  claim 25 , wherein said performing the pre-docking conformational search comprises: 
 randomly generating a plurality of uniformly distributed conformations of the ligand;    minimizing a strain and a solvent accessible surface area of each conformation of the plurality of uniformly distributed conformations;    using the strain of each conformation to rank the conformations; and    clustering the conformations and retaining a desired number of top clusters of conformations, said retained number of top clusters of conformations comprising said multiple solution conformations of the ligand.    
     
     
         29 . The at least one program storage device of  claim 25 , wherein said generating the binding site image includes at least one of creating a list of apolar hot spots identifying points in the binding site that are favorable for an apolar atom to bind, and generating a list of polar hot spots identifying points in the binding site that are favorable for a hydrogen bond donor or acceptor to bind.  
     
     
         30 . The device of  claim 25 , wherein said performing the pre-docking conformational search comprises: 
 randomly generating a plurality of uniformly distributed conformations of the ligand;    using a three-dimensional descriptor for each conformation to distinguish potentially active conformations from inactive conformations, and retaining potentially active conformations;    minimizing a strain of each potentially active conformation;    using the strain and a solvent accessible surface area of each potentially active conformation to rank the potentially active conformations; and    clustering the conformations and retaining a desired number of top clusters of conformations, said retained number of top clusters of conformations comprising said multiple solution conformations of the ligand.    
     
     
         31 . The device of  claim 30 , wherein said three-dimensional descriptor is selected from the group consisting of a polar solvent accessible surface area, an apolar solvent accessible surface area, number of internal interactions and radius of gyration.  
     
     
         32 . The at least one program storage device of  claim 29 , wherein said generating the binding site image further comprises: 
 placing a grid around the binding site of the target molecule;    determining a hot spot search volume using said grid;    determining hot spots using a grid-like search of the hot spot search volume; and    for each type of hot spot, clustering the hot spots and retaining a desired number of clusters of hot spots with best scores, said desired number of clusters comprising said multiple hot spots to be employed by said matching.    
     
     
         33 . The at least one program storage device of  claim 25 , wherein said matching comprises: 
 matching atoms of the at least one solution conformation to appropriate hot spots of the target molecule by positioning the at least one solution conformation as a rigid body into the binding site image;    defining a match, said match determining a unique rigid body transformation; and    using the unique rigid body transformation to place the at least one solution conformation of the ligand into the binding site of the target molecule.    
     
     
         34 . The at least one program storage device of  claim 33 , wherein said determining the unique rigid body transformation comprises determining the unique rigid body transformation that minimizes:  
       
         
           
             
               
                 I 
                  
                 
                   ( 
                   
                     R 
                     , 
                     T 
                   
                   ) 
                 
               
               = 
               
                 
                   ∑ 
                   
                     j 
                     = 
                     1 
                   
                   3 
                 
                  
                 
                   
                      
                     
                       
                         H 
                         j 
                       
                       - 
                       
                         RA 
                         j 
                       
                       - 
                       T 
                     
                      
                   
                   2 
                 
               
             
           
           
           
               
           
         
         where: 
 H j =a j th  hot spot of the target molecule;  
 A j =a j th  atom of the at least one solution conformation;  
 R=a 3×3 rotation matrix; and  
 T=a translation vector.  
 
       
     
     
         35 . The at least one program storage device of  claim 25 , wherein said optimizing comprises optimizing multiple target molecule-ligand complex formations, said optimizing comprising: 
 eliminating each ligand position having a predetermined percentage of ligand atoms with a steric clash;    ranking remaining ligand positions using an atom pairwise score with a desired atom score cutoff;    after ranking, clustering the ligand positions and selecting a top number n of ligand positions; and    optimizing each ligand position of the n positions, allowing the translation, rotation and rotatable bonds of the ligand to vary.    
     
     
         36 . The at least one program storage device of  claim 25 , wherein said optimizing comprises optimizing each ligand position of the n positions using a BFGS optimization algorithm with a simple atom pairwise score, allowing the translation, rotation and rotatable bonds of the ligand to vary.  
     
     
         37 . A method of assessing a combinatorial library for complementarity to a target molecule having at least one binding site, said combinatorial library comprising a plurality of ligands, each based on a common core, said method comprising: 
 docking each ligand of the plurality of ligands to the target molecule to generate a plurality of ligand positions relative to the target molecule in a plurality of ligand-target molecule complex formations, said plurality of ligand positions comprising a plurality of common core positions relative to the target molecule;    determining an rms deviation of each common core position of said plurality of common core positions from other common core positions; and    forming clusters according to said rms deviation.    
     
     
         38 . A method according to  claim 37 , additionally comprising rating complementarity of the combinatorial library to the target molecule according to number of ligands in a cluster having a minimum rms deviation relative to number of ligands in the combinatorial library.  
     
     
         39 . A method according to  claim 37 , wherein said determining an rms deviation comprises: 
 placing a grid around a binding site of the target molecule;    for each ligand position, determining a location on the grid corresponding to the center of mass of the common core; and    determining the rms deviation of each common core position from every other common core position having a location on the grid within a predetermined distance.    
     
     
         40 . A method according to  claim 37  wherein said forming clusters comprises forming clusters using a single linkage clustering algorithm.  
     
     
         41 . A method according to  claim 37  wherein said docking each ligand comprises: 
 performing a pre-docking conformational search to generate multiple solution conformations of each ligand;  
 generating a binding site image of the target molecule, said binding site image comprising multiple hot spots;  
 matching hot spots of the binding site image to atoms in at least one solution conformation of the multiple solution conformations of each ligand to obtain at least one ligand position relative to the target molecule in a ligand-target molecule complex formation; and  
 optimizing the at least one ligand position while allowing translation, orientation and rotatable bonds of the ligand to vary, and while holding the target molecule fixed.  
 
     
     
         42 . A method of comparing a plurality of combinatorial libraries for complementarity to a target molecule having at least one binding site, each of said plurality of combinatorial libraries comprising a plurality of ligands, each based on a common core, said method comprising: 
 for each combinatorial library, docking each ligand of the plurality of ligands to the target molecule to generate a plurality of ligand positions relative to the target molecule in a plurality of ligand-target molecule complex formations, said plurality of ligand positions comprising a plurality of common core positions relative to the target molecule;    determining an rms deviation of each common core position of said plurality of common core positions from other common core positions;    forming clusters according to said rms deviation; and    ranking the plurality of combinatorial libraries according to number of ligands in a top cluster of said clusters, relative to total number of ligands in each combinatorial library.    
     
     
         43 . A method according to  claim 42 , additionally comprising prioritizing high throughput screening of each combinatorial library for activity against a biotarget according to said ranking.  
     
     
         44 . A system for assessing a combinatorial library for complementarity to a target molecule having at least one binding site, said combinatorial library comprising a plurality of ligands, each based on a common core, said system comprising: 
 means for docking each ligand of the plurality of ligands to the target molecule to generate a plurality of ligand positions relative to the target molecule in a plurality of ligand-target molecule complex formations, said plurality of ligand positions comprising a plurality of common core positions relative to the target molecule;    means for determining an rms deviation of each common core position of said plurality of common core positions from other common core positions; and    means for forming clusters according to said rms deviation.    
     
     
         45 . A system according to  claim 44 , additionally comprising means for rating complementarity of the combinatorial library to the target molecule according to number of ligands in a cluster having a minimum rms deviation relative to number of ligands in the combinatorial library.  
     
     
         46 . A system according to  claim 44 , wherein said means for determining an rms deviation comprises: 
 means for placing a grid around a binding site of the target molecule;    means for, for each ligand position, determining a location on the grid corresponding to the center of mass of the common core; and    means for determining the rms deviation of each common core position from every other common core position having a location on the grid within a predetermined distance.    
     
     
         47 . A system according to  claim 44 , wherein said means for forming clusters comprises means for forming clusters using a single linkage clustering algorithm.  
     
     
         48 . A system according to  claim 44 , wherein said means for docking each ligand comprises: 
 means for performing a pre-docking conformational search to generate multiple solution conformations of each ligand;    means for generating a binding site image of the target molecule, said binding site image comprising multiple hot spots;    means for matching hot spots of the binding site image to atoms in at least one solution conformation of the multiple solution conformations of each ligand to obtain at least one ligand position relative to the target molecule in a ligand-target molecule complex formation; and    means for optimizing the at least one ligand position while allowing translation, orientation and rotatable bonds of the ligand to vary, and while holding the target molecule fixed.    
     
     
         49 . A system for comparing a plurality of combinatorial libraries for complementarity to a target molecule having at least one binding site, each of said plurality of combinatorial libraries comprising a plurality of ligands, each based on a common core, said method comprising: 
 for each combinatorial library, means for docking each ligand of the plurality of ligands to the target molecule to generate a plurality of ligand positions relative to the target molecule in a plurality of ligand-target molecule complex formations, said plurality of ligand positions comprising a plurality of common core positions relative to the target molecule;    means for determining an rms deviation of each common core position of said plurality of common core positions from other common core positions;    means for forming clusters according to said rms deviation; and    means for ranking the plurality of combinatorial libraries according to number of ligands in a top cluster of said clusters, relative to total number of ligands in each combinatorial library.    
     
     
         50 . A system according to  claim 49 , additionally comprising means for prioritizing high throughput screening of each combinatorial library for activity against a biotarget according to said ranking.  
     
     
         51 . At least one program storage device readable by a machine, tangibly embodying at least one program of instructions executable by the machine to perform a method for assessing a combinatorial library for complementarity to a target molecule having at least one binding site, said combinatorial library comprising a plurality of ligands, each based on a common core, said method comprising: 
 docking each ligand of the plurality of ligands to the target molecule to generate a plurality of ligand positions relative to the target molecule in a plurality of ligand-target molecule complex formations, said plurality of ligand positions comprising a plurality of common core positions relative to the target molecule;    determining an rms deviation of each common core position of said plurality of common core positions from other common core positions; and    forming clusters according to said rms deviation.    
     
     
         52 . The at least one program storage device according to  claim 51 , wherein said method additionally comprises rating complementarity of the combinatorial library to the target molecule according to number of ligands in a cluster having a minimum rms deviation relative to number of ligands in the combinatorial library.  
     
     
         53 . The at least one program storage device according to  claim 51 , wherein said determining an rms deviation comprises: 
 placing a grid around a binding site of the target molecule;    for each ligand position, determining a location on the grid corresponding to the center of mass of the common core; and    determining the rms deviation of each common core position from every other common core position having a location on the grid within a predetermined distance.    
     
     
         54 . The at least one program storage device according to  claim 51 , wherein said forming clusters comprises forming clusters using a single linkage clustering algorithm.  
     
     
         55 . The at least one program storage device according to  claim 51 , wherein said docking each ligand comprises: 
 performing a pre-docking conformational search to generate multiple solution conformations of each ligand;    generating a binding site image of the target molecule, said binding site image comprising multiple hot spots;    matching hot spots of the binding site image to atoms in at least one solution conformation of the multiple solution conformations of each ligand to obtain at least one ligand position relative to the target molecule in a ligand-target molecule complex formation; and    optimizing the at least one ligand position while allowing translation, orientation and rotatable bonds of the ligand to vary, and while holding the target molecule fixed.    
     
     
         56 . At least one program storage device readable by a machine, tangibly embodying at least one program of instructions executable by the machine to perform a method of comparing a plurality of combinatorial libraries for complementarity to a target molecule having at least one binding site, each of said plurality of combinatorial libraries comprising a plurality of ligands, each based on a common core, said method comprising: 
 for each combinatorial library, docking each ligand of the plurality of ligands to the target molecule to generate a plurality of ligand positions relative to the target molecule in a plurality of ligand-target molecule complex formations, said plurality of ligand positions comprising a plurality of common core positions relative to the target molecule;    determining an rms deviation of each common core position of said plurality of common core positions from other common core positions;    forming clusters according to said rms deviation; and    ranking the plurality of combinatorial libraries according to number of ligands in a top cluster of said clusters, relative to total number of ligands in each combinatorial library.    
     
     
         57 . The at least one program storage device of  claim 56 , additionally comprising prioritizing high throughput screening of each combinatorial library for activity against a biotarget according to said ranking.