System and method of automatically generating chemical compounds with desired properties

• US 5,463,564 A
• Filed: 09/16/1994
• Issued: 10/31/1995
• Est. Priority Date: 09/16/1994
• Status: Expired due to Term

• Current Assignee: 3-Dimensional Pharmaceuticals, Inc. (Johnson & Johnson)
• Original Assignee: 3-Dimensional Pharmaceuticals, Inc. (Johnson & Johnson)
• Inventors: Soll, Richard M., Bone, Roger F., Salemme, Francis R., Agrafiotis, Dimitris K.
• Primary Examiner(s): Voeltz, Emanuel T.
• Assistant Examiner(s): CHOI, KYLE JAEHUN
• Application Number:
  US08/306,915
• Time in Patent Office: 410 Days
• Field of Search: 364/496, 364/497, 364/499, 364/500, 436/43, 436/50, 436/55, 423/659, 424/2, 935/85-88
• US Class Current: 700/268
• CPC Class Codes: A61P 7/02 Antithrombotic agents; Anticoagulants; Platelet aggregation inhibitors
• CPC Class Codes: B01J 19/0046 Sequential or parallel reactions, e.g. for the synthesis of polypeptides or polynucleotides; Apparatus and devices for combinatorial chemistry or for making molecular arrays synthesis methods per se C40B50/00
• CPC Class Codes: B01J 2219/0059 Sequential processes
• CPC Class Codes: B01J 2219/00592 Split-and-pool, mix-and-divide processes
• CPC Class Codes: B01J 2219/00596 Solid-phase processes
• CPC Class Codes: B01J 2219/00599 Solution-phase processes
• CPC Class Codes: B01J 2219/00689 using computers
• CPC Class Codes: B01J 2219/00691 using robots
• CPC Class Codes: B01J 2219/00695 Synthesis control routines, e.g. using computer programs
• CPC Class Codes: B01J 2219/00698 Measurement and control of process parameters
• CPC Class Codes: B01J 2219/007 Simulation or vitual synthesis
• CPC Class Codes: B01J 2219/0072 Organic compounds
• CPC Class Codes: B01J 2219/00725 Peptides
• CPC Class Codes: C07K 1/047 Simultaneous synthesis of different peptide species; Peptide libraries
• CPC Class Codes: C40B 40/10 Libraries containing peptides or polypeptides, or derivatives thereof
• CPC Class Codes: C40B 50/08 Liquid phase synthesis, i.e. wherein all library building blocks are in liquid phase or in solution during library creation; Particular methods of cleavage from the liquid support
• CPC Class Codes: G16B 15/30 Drug targeting using structural data; Docking or binding prediction
• CPC Class Codes: G16B 35/00 ICT specially adapted for in silico combinatorial libraries of nucleic acids, proteins or peptides
• CPC Class Codes: G16C 20/10 Analysis or design of chemical reactions, syntheses or processes
• CPC Class Codes: G16C 20/30 Prediction of properties of chemical compounds, compositions or mixtures
• CPC Class Codes: G16C 20/60 : In silico combinatorial chemistry
• CPC Class Codes: G16C 20/62 : Design of libraries
• CPC Class Codes: G16C 20/70 : Machine learning, data mining or chemometrics
• CPC Class Codes: Y10T 436/11 : Automated chemical analysis

Abstract

A computer based, iterative process for generating chemical entities with defined physical, chemical and/or bioactive properties. During each iteration of the process, (1) a directed diversity chemical library is robotically generated in accordance with robotic synthesis instructions; (2) the compounds in the directed diversity chemical library are analyzed to identify compounds with the desired properties; (3) structure-property data are used to select compounds to be synthesized in the next iteration; and (4) new robotic synthesis instructions are automatically generated to control the synthesis of the directed diversity chemical library for the next iteration.

Claims

1. A computer based method of automatically generating compounds having a prescribed set of properties, comprising the steps of:

(1) robotically synthesizing, in accordance with robotic synthesis instructions, a directed diversity chemical library comprising a plurality of chemical compounds;

(2) robotically analyzing said chemical compounds to obtain structure-activity data pertaining thereto;

(3) comparing, under computer control, said structure-activity data of said chemical compounds against said prescribed set of properties to identify any of said chemical compounds substantially conforming to said prescribed set of properties;

(4) classifying, under computer control, said identified chemical compounds as lead compounds;

(5) analyzing, under computer control, said structure-activity data of said compounds and historical structure-activity data pertaining to compounds synthesized and analyzed in the past to derive structure-activity models having enhanced predictive and discriminating capabilities;

(6) identifying, under computer control and in accordance with said structure-activity models, reagents from a reagent database that, when combined, will produce a set of compounds predicted to exhibit activity/properties more closely matching said prescribed set of properties;

(7) generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said set of compounds; and

(8) repeating steps (1)-(7), wherein step (1) is repeated using said generated robotic synthesis instructions.

2. The method of claim 1, wherein step (6) comprises the step of:

identifying, under computer control and in accordance with said structure-activity models, reagents from a reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to validate said structure-activity models, wherein said first and second sets of compounds are not mutually exclusive;

wherein step (7) comprises the step of generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second set of compounds.

3. The method of claim 1, wherein step (6) comprises the step of:

identifying, under computer control and in accordance with said structure-activity models, reagents from a reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to discriminate between said structure-activity models, wherein said first and second sets of compounds are not mutually exclusive;

wherein step (7) comprises the step of generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second set of compounds.

4. The method of claim 1, wherein step (6) comprises the step of:

identifying, under computer control and in accordance with said structure-activity models, reagents from a reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to validate said structure-activity models, and a third set of compounds predicted to have a superior ability to discriminate between said structure-activity models, wherein said first, second, and third sets of compounds are not mutually exclusive;

wherein step (7) comprises the step of generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second and third set of compounds.

5. The method of claim 1, wherein step (6) comprises the step of:

identifying, under computer control and in accordance with said structure-activity models, reagents from a reagent database that, when combined, will produce a second set of compounds predicted to have superior three-dimensional receptor fit, wherein said first and second sets of compounds are not mutually exclusive;

wherein step (7) comprises the step of generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second set of compounds.

6. The method of claim 1, wherein step (6) comprises the step of:

identifying under computer control reagents from said reagent database that, when combined, will produce a second set of compounds with structural, physical, or chemical characteristics similar to characteristics of compounds in a structure-activity database whose activity/properties most closely conform to said prescribed set of activity/properties.

7. The method of claim 1, wherein step (6) comprises the steps of:

(a) generating a list of potential chemical compounds to possibly synthesize, each of said potential chemical compounds synthesized using N of said identified reagents; and

(b) selecting from said list of potential chemical compounds a plurality of candidate compounds that are suitable for synthesis.

8. The method of claim 7, wherein N is greater or equal to 1 and less than 500.

9. The method of claim 7, wherein N is of a magnitude such that each compound synthesized contains three variable subunits.

10. The method of claim 7, wherein step (2) comprises the steps of:

analyzing said chemical compounds to obtain structure and composition data pertaining thereto;

analyzing said structure and composition data to generate chemical synthesis indicia indicating which of said chemical compounds were adequately synthesized, and which of said chemical compounds were not adequately synthesized; and

storing said structure and composition data and said chemical synthesis indicia in a structure-activity database, said structure-activity database also having stored therein structure and composition data and chemical synthesis indicia pertaining to previously synthesized chemical compounds.

11. The method of claim 10, wherein step (b) comprises the steps of:

retrieving from said structure-activity database any chemical synthesis indicia pertaining to said potential chemical compounds; and

selecting as candidate compounds any of said potential chemical compounds that were previously adequately synthesized as indicated by said retrieved chemical synthesis indicia, if any, respectively associated with said potential chemical compounds.

12. The method of claim 10, wherein step (b) comprises the steps of:

retrieving from said structure-activity database any chemical synthesis indicia pertaining to said potential chemical compounds; and

eliminating from consideration as candidate compounds any of said potential chemical compounds that were previously inadequately synthesized as indicated by said retrieved chemical synthesis indicia, if any, respectively associated with said potential chemical compounds.

13. The method of claim 7, wherein step (7) comprises the step of:

generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said candidate compounds.

14. The method of claim 7, further comprising the step of:

(c) selecting an optimal set of said candidate compounds to synthesize based on at least one of the following factors;

(i) their respective predicted abilities to exhibit activity/properties more closely matching said prescribed set of activity/properties as indicated by said structure-activity models;

(ii) their respective predicted abilities to validate said structure-activity models;

(iii) their respective predicted abilities to discriminate between said structure-activity models;

(iv) their respective predicted abilities to have superior three-dimensional receptor fit; and

(v) similarity between their respective structural, physical, or chemical characteristics and characteristics of compounds in a structure-activity database whose activity/properties most closely conform to said prescribed set of activity/properties.

15. The method of claim 14, wherein step (c) comprises the step of:

selecting said optimal set by individually ranking said candidate compounds based on at least one of factors (i)-(v).

16. The method of claim 14, wherein step (c) comprises the step of:

selecting said optimal set by ranking combinations of said candidate compounds based on at least one of factors (i)-(v).

17. The method of claim 14, wherein step (7) comprises the step of:

generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said optimal set of candidate compounds.

18. The method of claim 1, wherein step (6) is performed according to operator input.

19. The method of claim 1, wherein step (7) comprises the steps of:

receiving operator input pertaining to the generation of said robotic synthesis instructions; and

generating said robotic synthesis instructions based, at least in part, on said operator input.

20. The method of claim 1, wherein said reagents include amines suitable for synthesizing thrombin inhibitors.

21. A computer-based system for automatically generating compounds having a prescribed set of activity/properties, comprising:

one or more chemical synthesis robots to robotically synthesize, in accordance with robotic synthesis instructions, a directed diversity chemical library comprising a plurality of chemical compounds;

one or more analysis robots to robotically analyze said chemical compounds to obtain structure-activity data pertaining thereto;

a synthesis protocol generator, comprising;

comparing means for comparing said structure-activity data of said chemical compounds against said prescribed set of activity/properties to identify any of said chemical compounds substantially conforming to said prescribed set of activity/properties;

classifying means for classifying said identified chemical compounds as lead compounds;

structure-activity model derivation means for analyzing said structure-activity data of said compounds and historical structure-activity data pertaining to compounds synthesized and analyzed in the past to derive structure-activity models having enhanced predictive and discriminating capabilities;

reagent identifying means for identifying, in accordance with said structure-activity models, reagents from a reagent database that, when combined, will produce a set of compounds predicted to exhibit activity/properties more closely matching said prescribed set of activity/properties; and

robotic synthesis instruction generating means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robots to robotically synthesize said set of compounds.

22. The system of claim 21, wherein said reagent identifying means comprises:

means for identifying, in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to validate said structure-activity models, wherein said first and second sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robots to robotically synthesize said second set of compounds.

23. The system of claim 21, wherein said reagent identifying means comprises:

means for identifying, in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to discriminate between said structure-activity models, wherein said first and second sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robots to robotically synthesize said second set of compounds.

24. The system of claim 21, wherein said reagent identifying means comprises:

means for identifying, in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to validate said structure-activity models, and a third set of compounds predicted to have a superior ability to discriminate between said structure-activity models, wherein said first, second, and third sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robots to robotically synthesize said second and third set of compounds.

25. The system of claim 21, wherein said reagent identifying means comprises:

means for identifying, in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have superior three-dimensional receptor fit, wherein said first and second sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robots to robotically synthesize said second set of compounds.

26. The system of claim 21, wherein said reagent identifying means comprises:

means for identifying reagents from said reagent database that, when combined, will produce a second set of compounds with structural, physical, or chemical characteristics similar to characteristics of compounds in a structure-activity database whose activity/properties most closely conform to said prescribed set of activity/properties, wherein said first and second sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robots to robotically synthesize said second set of compounds.

27. The system of claim 21, wherein said synthesis protocol generator further comprises:

means for generating a list of potential chemical compounds to possibly synthesize, each of said potential chemical compounds synthesized using N of said reagents identified by said reagent identifying means; and

candidate compound identifying means for selecting from said list of potential chemical compounds a plurality of candidate compounds that are suitable for synthesis.

28. The system of claim 27, wherein N is of a magnitude such that each compound synthesized contains three variable subunits.

29. The method of claim 27, wherein N is greater or equal to 1 and less than 500.

30. The system of claim 27, wherein said analysis robots comprise:

means for analyzing said chemical compounds to obtain structure and composition data pertaining thereto;

means for analyzing said structure and composition data to generate chemical synthesis indicia indicating which of said chemical compounds were adequately synthesized, and which of said chemical compounds were not adequately synthesized; and

means for storing said structure and composition data and said chemical synthesis indicia in a structure-activity database, said structure-activity database also having stored therein structure and composition data and chemical synthesis indicia pertaining to previously synthesized chemical compounds.

31. The system of claim 30, wherein said candidate compound identifying means comprises:

means for retrieving from said structure-activity database any chemical synthesis indicia pertaining to said potential chemical compounds; and

means for selecting as candidate compounds any of said potential chemical compounds that were previously adequately synthesized as indicated by said retrieved chemical synthesis indicia, if any, respectively associated with said potential chemical compounds.

32. The system of claim 27, wherein said robotic synthesis instruction generating means comprises:

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robots to robotically synthesize said candidate compounds.

33. The system of claim 27, wherein said synthesis protocol generator further comprises:

optimal set selecting means for selecting an optimal set of said candidate compounds to synthesize based on at least one of the following factors;

(i) their respective predicted abilities to exhibit activity/properties more closely matching said prescribed set of activity/properties as indicated by said structure-activity models;

(ii) their respective predicted abilities to validate said structure-activity models;

(iii) their respective predicted abilities to discriminate between said structure-activity models;

(iv) their respective predicted abilities to have superior three-dimensional receptor fit; and

(v) similarity between their respective structural, physical, or chemical characteristics and characteristics of compounds in a structure-activity database whose activity/properties most closely conform to said prescribed set of activity/properties.

34. The system of claim 33, wherein said optimal set selecting means comprises:

means for selecting said optimal set by individually ranking said candidate compounds based on at least one of factors (i)-(v).

35. The system of claim 33, wherein said optimal set selecting means comprises:

means for selecting said optimal set by ranking combinations of said candidate compounds based on at least one of factors (i)-(v).

36. The system of claim 33, wherein said robotic synthesis instruction generating means comprises:

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robots to robotically synthesize said optimal set of candidate compounds.

37. The system of claim 21, wherein said reagent identifying means operates according to operator input.

38. The system of claim 21, wherein said robotic synthesis instruction generating means comprises:

means for receiving operator input pertaining to the generation of said robotic synthesis instructions; and

means for generating said robotic synthesis instructions based, at least in part, on said operator input.

39. The system of claim 21, wherein said reagents include amines suitable for synthesizing thrombin inhibitors.

40. A computer based method of iteratively generating a plurality of directed diversity chemical libraries each comprising multiple chemical compounds, wherein chemical compounds in said directed diversity chemical libraries more closely conform to a prescribed set of activity/properties during each successive iteration, the method comprising the steps of:

(1) robotically synthesizing, in accordance with robotic synthesis instructions, a directed diversity chemical library comprising a plurality of chemical compounds;

(2) robotically analyzing said chemical compounds to obtain structure-activity data pertaining thereto;

(3) analyzing, under computer control, said structure-activity data of said compounds and historical structure-activity data pertaining to compounds synthesized and analyzed during prior iterations to derive structure-activity models having enhanced predictive and discriminating capabilities;

(4) identifying, under computer control and in accordance with said structure-activity models, reagents from a reagent database that, when combined, will produce a set of compounds predicted to exhibit activity/properties more closely matching said prescribed set of activity/properties;

(5) generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said set of compounds; and

(6) repeating steps (1)-(5), wherein step (1) is repeated using said generated robotic synthesis instructions.

41. The method of claim 40, wherein step (4) comprises the step of:

identifying, under computer control and in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to validate said structure-activity models, wherein said first and second sets of compounds are not mutually exclusive;

wherein step (5) comprises the step of generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second set of compounds.

42. The method of claim 40, wherein step (4) comprises the step of:

identifying, under computer control and in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to discriminate between said structure-activity models, wherein said first and second sets of compounds are not mutually exclusive;

wherein step (5) comprises the step of generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second set of compounds.

43. The method of claim 40, wherein step (4) comprises the step of:

identifying, under computer control and in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to validate said structure-activity models, and a third set of compounds predicted to have a superior ability to discriminate between said structure-activity models, wherein said first, second, and third sets of compounds are not mutually exclusive;

wherein step (5) comprises the step of generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second and third set of compounds.

44. The method of claim 40, wherein step (4) comprises the step of:

identifying, under computer control and in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have superior three-dimensional receptor fit, wherein said first and second sets of compounds are not mutually exclusive;

wherein step (5) comprises the step of generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second set of compounds.

45. The method of claim 40, wherein step (4) comprises the step of:

identifying under computer control reagents from said reagent database that, when combined, will produce a second set of compounds with structural, physical, or chemical characteristics similar to characteristics of compounds in a structure-activity database whose activity/properties most closely conform to said prescribed set of activity/properties, wherein said first and second sets of compounds are not mutually exclusive;

wherein step (5) comprises the step of;

generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second set of compounds.

46. The method of claim 40, further comprising the following steps which are performed between steps (4) and (5):

(a) generating a list of potential chemical compounds to possibly synthesize, each of said potential chemical compounds synthesized using N of said identified reagents; and

(b) selecting from said list of potential chemical compounds a plurality of candidate compounds that are suitable for synthesis.

47. The method of claim 46, wherein step (5) comprises the step of:

generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said candidate compounds.

48. The method of claim 46, further comprising the step of:

(c) selecting an optimal set of said candidate compounds to synthesize based on at least one of the following factors;

(i) their respective predicted abilities to exhibit activity/properties more closely matching said prescribed set of activity/properties as indicated by said structure-activity models;

(ii) their respective predicted abilities to validate said structure-activity models;

(iii) their respective predicted abilities to discriminate between said structure-activity models;

(iv) their respective predicted abilities to have superior three-dimensional receptor fit; and

(v) similarity between their respective structural, physical, or chemical characteristics and characteristics of compounds in a structure-activity database whose activity/properties most closely conform to said prescribed set of activity/properties.

49. The method of claim 48, wherein step (c) comprises the step of:

selecting said optimal set by individually ranking said candidate compounds based on at least one of factors (i)-(v).

50. The method of claim 48, wherein step (c) comprises the step of:

selecting said optimal set by ranking combinations of said candidate compounds based on at least one of factors (i)-(v).

51. The method of claim 48, wherein step (5) comprises the step of:

generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said optimal set of candidate compounds.

52. A synthesis protocol generator for use in a system that automatically generates compounds having a prescribed set of activity/properties, said system comprising at least one chemical synthesis robot to robotically synthesize, in accordance with robotic synthesis instructions, a directed diversity chemical library comprising a plurality of chemical compounds, and at least one analysis robot to robotically analyze said chemical compounds to obtain structure-activity data pertaining thereto, said synthesis protocol generator comprising:

structure-activity model derivation means for analyzing said structure-activity data of said compounds and historical structure-activity data pertaining to compounds synthesized and analyzed in the past to derive structure-activity models having enhanced predictive and discriminating capabilities;

reagent identifying means for identifying, in accordance with said structure-activity models, reagents from a reagent database that, when combined, will produce a set of compounds predicted to exhibit activity/properties more closely matching said prescribed set of activity/properties; and

robotic synthesis instruction generating means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robot to robotically synthesize said set of compounds.

53. The synthesis protocol generator of claim 52, further comprising:

comparing means for comparing said structure-activity data of said chemical compounds against said prescribed set of activity/properties to identify any of said chemical compounds substantially conforming to said prescribed set of activity/properties; and

classifying means for classifying said identified chemical compounds as lead compounds.

54. The synthesis protocol generator of claim 52, wherein said reagent identifying means comprises:

means for identifying, in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to validate said structure-activity models, wherein said first and second sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robot to robotically synthesize said second set of compounds.

55. The synthesis protocol generator of claim 52, wherein said reagent identifying means comprises:

means for identifying, in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to discriminate between said structure-activity models, wherein said first and second sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robot to robotically synthesize said second set of compounds.

56. The synthesis protocol generator of claim 52, wherein said reagent identifying means comprises:

means for identifying, in accordance with said structure-activity models, reagent from said reagent database that, when combined, will produce a second set of compounds predicted to have a superior ability to validate said structure-activity models, and a third set of compounds predicted to have a superior ability to discriminate between said structure-activity models, wherein said first, second, and third sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robot to robotically synthesize said second and third set of compounds.

57. The synthesis protocol generator of claim 52, wherein said reagent identifying means comprises:

means for identifying, in accordance with said structure-activity models, reagents from said reagent database that, when combined, will produce a second set of compounds predicted to have superior three-dimensional receptor fit, wherein said first and second sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robot to robotically synthesize said second set of compounds.

58. The synthesis protocol generator of claim 52, wherein said reagent identifying means comprises:

means for identifying under computer control reagents from said reagent database that, when combined, will produce a second set of compounds with structural, physical, or chemical characteristics similar to characteristics of compounds in a structure-activity database whose activity/properties most closely conform to said prescribed set of activity/properties, wherein said first and second sets of compounds are not mutually exclusive;

wherein said robotic synthesis instruction generating means comprises;

means for generating, under computer control, robotic synthesis instructions that, when executed, enable robotic synthesis of said second set of compounds.

59. The synthesis protocol generator of claim 52, further comprising:

means for generating a list of potential chemical compounds to possibly synthesize, each of said potential chemical compounds comprising N of said reagents identified by said reagent identifying means; and

candidate compound identifying means for selecting from said list of potential chemical compounds a plurality of candidate compounds that are suitable for synthesis.

60. The synthesis protocol generator of claim 59, wherein said robotic synthesis instruction generating means comprises:

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robot to robotically synthesize said candidate compounds.

61. The synthesis protocol generator of claim 59, wherein said synthesis protocol generator further comprises:

optimal set selecting means for selecting an optimal set of said candidate compounds to synthesize based on at least one of the following factors;

(i) their respective predicted abilities to exhibit activity/properties more closely matching said prescribed set of activity/properties as indicated by said structure-activity models;

(ii) their respective predicted abilities to validate said structure-activity models;

(iii) their respective predicted abilities to discriminate between said structure-activity models;

(iv) their respective predicted abilities to have superior three-dimensional receptor fit; and

(v) similarity between their respective structural, physical, or chemical characteristics and characteristics of compounds in a structure-activity database whose activity/properties most closely conform to said prescribed set of activity/properties.

62. The synthesis protocol generator of claim 61, wherein said optimal set selecting means comprises:

means for selecting said optimal set by individually ranking said candidate compounds based on at least one of factors (i)-(v).

63. The synthesis protocol generator of claim 61, wherein said optimal set selecting means comprises:

means for selecting said optimal set by ranking combinations of said candidate compounds based on at least one of factors (i)-(v).

64. The synthesis protocol generator of claim 61, wherein said robotic synthesis instruction generating means comprises:

means for generating robotic synthesis instructions that, when executed, enable said chemical synthesis robot to robotically synthesize said optimal set of candidate compounds.