Design of molecules

Abstract

A method for computational drug design using an evolutionary algorithm, comprises evaluating virtual molecules according to vector distance (VD) to at least one achievement objective that defines a desired ideal molecule. In one method the invention comprises defining a set of n achievement objectives (OA1-n), where n is at least one; defining a population (PG=0) of at least one molecule; selecting an initial population (Pparent) of at least one molecule (I1-In) from the population (PG=0); and evaluating members (I1-In) of the initial population (Pparent) against at least one of the n achievement objectives (OA1-x), where x is from 1 to n.

1 . A method for designing a drug compound having a particular property for a desired use, the method comprising:
 defining a set of n achievement objectives (O A   1-n ), where n is at least one;   defining a population (P G=0 ) of at least one molecule;   selecting an initial population (P parent )of at least one molecule (I 1 -I n ) from the population (P G=0 ); and   evaluating members (I 1 -I n ) of the initial population (P parent ) against at least one of the n achievement objectives (O A   1-x ), where x is from 1 to n;   wherein the evaluating comprises the calculation of a linear distance (VD) from the member (I 1 -I n ) to the at least one achievement objective (O A   1-x );   determining whether a stop condition is satisfied;   upon determining that a stop condition is not satisfied generating further populations (P G ; P G+1 ) of molecules and evaluating each one by an iterative process until a predefined stop condition is satisfied;   upon determining that a stop condition is satisfied:
 ranking members (I 1 -I n ) of the initial or further evaluated population (P parent  P G ; P G+1 ) according to linear distance (VD) and optionally Pareto frontier to the set of n achievement objectives (O A   1-n ); and 
 identifying at least the first ranked member (I 1 -I n ) of the evaluated initial or further population (P parent  P G ; P G+1 t). 
   
     
     
         2 . The method of  claim 1 , wherein the linear distance is a vector distance, optionally wherein the linear distance is the vector distance (VD) between each member (I 1 -I m ) and the at least one achievement objective (OA1-x) is defined as: 
       
         
           
             
               
                 V 
                 A 
               
               = 
               
                 
                   
                     ∑ 
                     
                       p 
                       = 
                       1 
                     
                     n 
                   
                    
                   
                       
                   
                    
                   
                     
                       ( 
                       
                         
                           x 
                           I 
                           p 
                         
                         - 
                         
                           x 
                           A 
                           p 
                         
                       
                       ) 
                     
                     2 
                   
                 
               
             
           
         
         wherein a desired value of each of the achievement objectives (O A   1-n ) has coordinates (x I   1 , . . . , x I   n ), and a calculated value of a parameter for each member (I 1 -I m ) of the initial population as evaluated against each of the achievement objectives (O A   1-n ) has coordinates (x A   1 , . . . , x A   n ). 
       
     
     
         3 . The method of  claim 1 , wherein the evaluating comprises:
 calculating the parameters of each member (I 1 -I n ) of the initial population (P parent ) for each of the at least one achievement objective (O A   1-x ) or each of the n achievement objectives (O A   1-n ); and   calculating the linear distance (VD) of each member (I 1 -I n ) of the initial population (P parent ) to the at least one achievement objective (O A   1-x ) or to each of the n achievement objectives (O A   1-n ); optionally calculating and assigning a Pareto frontier ranking to members (I 1 -I n ) of the initial population (P parent ); and   wherein the first ranked member has the shortest linear distance (VD) to the at least one achievement objective (O A   1-x ) or each of the n achievement objectives (O A   1-n ).   
     
     
         4 . The method of  claim 1 , wherein generating further populations (P G ; P G+1 ) of molecules and evaluating each one by an iterative process until a predefined stop condition is satisfied comprises, performing a first iteration (G=1) of an evolutionary algorithm to generate and evaluate a new population (P G ) of at least one molecule, the method comprising:
 transforming at least one member of the parent population (P parent ) to generate a transformed population (P transformed ) of at least one molecule;   defining a new population (P G ) of at least one molecule, the new population (P G ) comprising at least one member of the transformed population (P transformed );   optionally evaluating the population (P G ) against at least one achievement objective (O A   1-x ), and selecting molecules from the evaluated population (P G ) by applying a strategy function (S);   defining a new population (P G+1 ) of at least one molecule (I 1 -I n ); and   evaluating members (I 1 -I n ) of the new population (P G+1 ) against the at least one achievement objective (O A   1-x );   wherein the evaluating comprises the calculation of linear distance (VD) to the at least one achievement objective (O A   1-x ) and optionally a Pareto frontier for the members (I 1 -I n ) of the new population (P G+1 ).   
     
     
         5 . The method of  claim 4 , wherein the evaluating comprises:
 calculating the parameters of each member (I 1 -I n ) of the new population (P G+1 ) for each of the at least one achievement objective (O A   1-x ) or for each of the n achievement objectives (O A   1-n ); and   calculating the linear distance (VD) of each member (I 1 -I n ) of the new population (P G+1 ) to the at least one achievement objective (O A   1-x ) or to each of the n achievement objectives (O A   1-n ); optionally calculating and assigning a Pareto frontier ranking to members (I 1 -I n ) of the new population (P G+1 ).   
     
     
         6 . The method of  claim 4 , further comprising:
 determining whether a stop condition is satisfied, and if the stop condition is not satisfied:
 defining the evaluated new population (P G+1 ) as a new parent population (P parent ) of at least one molecule (I 1 -I n ); and 
 performing a second iteration (G=2) of the evolutionary algorithm by repeating the steps of  claim 5 . 
   
     
     
         7 . The method of  claim 6 , which further comprises, if the stop condition is satisfied:
 ranking members (I 1 -I n ) of the new population (P G+1 ) according to linear distance (VD) and optionally Pareto frontier to the set of n achievement objectives (O A   1-n ); and   identifying at least the first ranked member (I 1 -I n ) of the evaluated new population (P G+1 ), optionally wherein the first ranked member has the shortest vector distance (VD) to the achievement objectives (O A   1-n ).   
     
     
         8 . The method of  claim 4 , wherein P G =P parent +P transtormed . 
     
     
         9 . The method of  claim 4 , further comprising:
 applying at least one filter (F) to remove molecules that fail at least one predefined criteria of the filter (F), optionally wherein the at least one filter (F) is applied to molecules of the population (P G ) before evaluating the population (P G ) against at least one achievement objective (O A   1-x ).   
     
     
         10 . The method of  claim 7 , further comprising applying at least one filter (F) to remove molecules that fail at least one predefined criteria of the filter (F), wherein the filter (F) is applied to molecules of the population (P G+1 ) before ranking members (I 1 -I n ) of the new population (P G+1 ). 
     
     
         11 . The method of  claim 9 , wherein the at least one predefined criteria of the at least one filter (F) is selected from at least one of: non-broken molecule requirement; solubility; drug-like properties, such as absorption, distribution, metabolism, and excretion (ADME); molecular weight; hydrogen bonding capacity;
 octanol-water partition coefficient; toxicity; unwanted group definition; wanted group definition; total polar surface area; number of rotatable bonds; molecule size, such as number of atoms, number of rings, size of ring systems; number of functional groups, such as H-bond donors, H-bond acceptors; number of heteroatoms; and removing of duplicate molecules from the population (P G ).   
     
     
         12 . The method of  claim 4 , wherein optionally evaluating the population (P G ) against at least one achievement objective (O A   1-x ), and selecting molecules from the evaluated population (P G ) by applying a strategy function (S), comprises:
 identifying at least one desired activity (A) of an optimised molecule and defining a strategy function (S) to score each member (I 1 -I n ) of the population (P G ) against one or more of the at least one desired activity (A 1-n );   calculating the parameters of each member (I 1 -I n ) of the population (P G ) for at least one of the achievement objectives (O A   1-x ) relevant to the one or more desired activity (A 1-n );   determining the predicted activity (Prediction 1 to Prediction n) of each member (I 1 -I n ) of the population (P G ) for the one or more desired activity (A 1-n );   selecting the sub-population (P elite ) of molecules of the population (P G ) that satisfy the strategy function (S); and   optionally selecting a sub-population (P random ) of at least one molecule from the sub-population (P non-elite ) of molecules that do not satisfy the strategy function (S).   
     
     
         13 . The method of  claim 12 , wherein the new population (P G+1 ) of at least one molecule (I 1 -I n ) comprises P elite  or P elite  +P random . 
     
     
         14 . The method of  claim 12 , wherein the strategy function (S) is satisfied for molecules of the population (PG) that satisfy one of:
 the molecule is part of a subset of the population P G  where each molecule in the subset has a higher predicted activity (Prediction 1) than each molecule in the remainder of the population P G ; and   the predicted activity (Prediction 1) is greater than the sum of the mean predicted activity ([Prediction 1]Mean) and the standard deviation of the predicted activity ([Prediction 1]StdDev) for all members (I 1 -I n ) of the population (P G ); i.e. where:
   Prediction 1>[Prediction 1]Mean+[Prediction 1]StdDev 
   
     
     
         15 . The method of  claim 12 , wherein the at least one desired activity (A) of an optimised molecule is selected from one or more of: predicted activity against one or more target molecule (e.g. specificity, binding affinity, inhibition constant); predicted relative activity against one target molecule compared to another molecule; predicted selectivity for one or more target molecule over another molecule; predicted relative selectivity for more than one target molecule; predicted drug-like properties/scores (e.g. ADME); prioritisation of one or more ADME property; prioritisation of drug-like properties over one or more activity or specificity; and prioritisation of vector optimisation over Pareto frontier. 
     
     
         16 . The method of  claim 4 , wherein the transformations are derived from one or both of a database of known chemical transformations and a library of genetic algorithm operators. 
     
     
         17 . The method of  claim 1 , wherein the initial population (P parent ) of at least one molecule is one of:
 larger population (P G=0 ) of molecules based on at least one predetermined selection criteria; a population of molecules selected using one or more selection criteria selected from at least one of: 3D virtual docking, chemical similarity, database searching, Bayesian activity modelling, and an algorithm; or   a population of one molecule.   
     
     
         18 . The method of  claim 1 , wherein the set of n achievement objectives (O A   1-n ) comprises a plurality of parameter values that define properties of a desired optimised molecule and includes at least one of: inhibition activity against a target molecule; binding affinity to a target molecule; specificity for a target molecule; selectivity for the target molecule over a non-target molecule; pharmacokinetic properties; ADME scores; desirability scores; ligand efficiency; and parameters relating to interactions of an optimised molecule with two or more different target molecules or target sites. 
     
     
         19 . The method of  claim 1 , wherein the stop condition is selected from: the number of iterations (G=n) of the evolutionary algorithm; a predefined vector distance (VD) of a predefined number or proportion of molecules in a population (P parent ; P G ; P G+1 ) to one or more of the set of n achievement objectives (O A   1-n ); mean vector distance ([VD]Mean) of a population (P parent ; P G ; P G+1 ) of molecules to one or more of the set of n achievement objectives (O A   1-n ); the rate of change in the mean vector distance ([VD]Mean) of successive evaluated populations (P parent ; P G ; P G+1 ) of molecules; the rate of change in any other evaluated predefined criteria between successive populations (P parent ; P G ; P G+1 ) of molecules; molecular complexity; and a time limitation. 
     
     
         20 . The method of  claim 1 , further comprising:
 selecting at least one molecule of the population (P parent ) or (P G+1 ) based on its ranking;   identifying the at least one selected molecule as an optimised molecule; and optionally   synthesising the at least one selected molecule.