Methods for characterizing tissue or organ condition or status

• US 20050143628A1
• Filed: 06/17/2004
• Published: 06/30/2005
• Est. Priority Date: 06/18/2003
• Status: Abandoned Application

• Current Assignee: Merck Sharp & Dohme Corporation (Merck & Co., Inc.)
• Original Assignee: Merck Sharp & Dohme Corporation (Merck & Co., Inc.)
• Inventors: He, Yudong, Dai, Hongyue, Dai, Xudong, Ulrich, Roger G., Waring, Jeffrey F.
• Application Number:
  US10/872,157
• Publication Number:
  US 20050143628A1
• Time in Patent Office: Days
• US Class Current: 600/300
• CPC Class Codes: A61B 5/00 Measuring for diagnostic purposes radiation diagnosis A61B6/00; diagnosis by ultrasonic, sonic or infrasonic waves A61B8/00; Identification of persons
• CPC Class Codes: A61B 5/416 the spleen
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: G01N 2500/00 Screening for compounds of potential therapeutic value
• CPC Class Codes: G01N 33/567 utilising isolate of tissue or organ as binding agent
• CPC Class Codes: G16B 40/00 ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
• CPC Class Codes: G16B 40/30 Unsupervised data analysis

Abstract

The invention provides methods for characterizing the condition or status of a tissue or organ in a multicellular organism, e.g., an animal, by combining a plurality of clinical measures are combined into a composite clinical score (CCS) and using such a CCS to represent the condition or status of the tissue or organ. The invention provides methods for predicting the condition or status of a tissue or organ in a multicellular organism, e.g., an animal, based on measurements of a set of cellular constituent markers, e.g., measured expression levels of a set of marker genes. The invention also provides methods for selecting the set of marker genes whose expression levels can be used in determining the CCS.

Claims

1. A method for characterizing the condition of a tissue or organ in an animal, comprising determining a composite clinical score of said tissue or organ, wherein said composite clinical score is determined based on a plurality of k clinical measures of said tissue or organ of said animal.

2. The method of claim 1, wherein each of said plurality of k clinical measures is a converted clinical measure represented as deviations from the respective normal value.

3. The method of claim 2, wherein each of said converted clinical measures is calculated according to the equation

4. The method of claim 3, wherein each of said plurality of k clinical measures is sigmoidal transformed according to the equation

5. The method of claim 4, wherein said composite clinical score is calculated according to the equation

6. The method of claim 1, wherein said condition of said tissue or organ is a disease condition.

7. The method of claim 6, wherein said disease condition is inflammation or damage.

8. The method of any one of claims 1-7, further comprising classifying said tissue or organ according to a predetermined threshold of said composite clinical score, wherein said tissue or organ is classified into one or the other category depending on if said composite clinical score is greater or smaller than said predetermined threshold.

9. The method of claim 4, wherein said organ is liver and said plurality of k clinical measures are selected from the group consisting of the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), the serum level of cholesterol (Chol), the serum level of gamma-glutamyltranspeptidase (GGT), the serum level of albumin, the serum level of globulins, and the prothrombin time.

10. The method of claim 9, wherein said plurality of k clinical measures consist of the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), and the serum level of cholesterol (Chol).

11. The method of claim 10, wherein said serum level of alanine aminotransferase (ALT) is sigmoidal transformed with c of 3 and said serum level of alkaline phosphatase (ALP), said serum level of total bilirubin (Tbil), and said serum level of cholesterol (Chol) are each sigmodal transformed with c of 1.

12. The method of claim 11, wherein said composite clinical score is a hepatotoxicity score HS calculated according to the equation

101. A computer system comprising

a processor, and

a memory coupled to said processor and encoding one or more programs, wherein said one or more programs cause the processor to carry out the method of any one of claims 1, 13, 26, 42, and 57.

102. A computer program product for use in conjunction with a computer having a processor and a memory connected to the processor, said computer program product comprising a computer readable storage medium having a computer program mechanism encoded thereon, wherein said computer program mechanism may be loaded into the memory of said computer and cause said computer to carry out the method of any one of claims 1, 13, 26, 42, and 57.

13. A method for characterizing the condition of a tissue or organ in an animal, comprising determining a composite clinical score of said tissue or organ based on a cellular constituent profile of said tissue or organ, wherein said cellular constituent profile comprises measurements of a plurality of cellular constituents in cells of said tissue or organ.

14. The method of claim 13, wherein said composite clinical score of said tissue or organ is determined by a model estimator according to equation

CCS=f(z₁, z₂, . . . z_n)

where {z₁, z₂, . . . , z_n} are data characterizing said cellular constituent profile.

15. The method of claim 14, wherein said {z₁, z₂, . . . , z_n} are data in a feature space.

16. The method of claim 15, wherein said {z₁, z₂, . . . , z_n} are obtained by transforming said cellular constituent profile using a wavelet transformation of a suitable level.

17. The method of claim 16, wherein said wavelet transformation is a transformation using Daubechies wavelet.

18. The method of claim 14, wherein said model estimator is a neural network model.

19. A computer program encoding a model estimator for characterizing a condition of a tissue or organ in an animal, said computer program accepting data characterizing a cellular constituent profile of said tissue or organ, wherein said cellular constituent profile comprises measurements of a plurality of cellular constituent in cells of said tissue or organ, and outputting a composite clinical score of said tissue or organ, wherein said composite clinical score indicates said condition of said tissue or organ of said animal.

20. The computer program of claim 19, wherein said condition results from a perturbation to said tissue or organ.

21. The computer program of claim 19 or 20, wherein said data characterizing said cellular constituent profile are data in a feature space.

22. The computer program of claim 21, wherein said data in said feature space are obtained by transforming said cellular constituent profile using a wavelet transformation of a suitable level.

23. The computer program of claim 22, wherein said wavelet transformation is a transformation using Daubechies wavelets of a suitable level.

24. The computer program of claim 23, wherein said model estimator is a neural network model.

25. The computer program of claim 20, wherein said perturbation is a drug perturbation and wherein said condition results from the toxicity of said drug.

26. A method for evaluating the toxicity of a drug to a tissue or organ in an animal, comprising determining a composite clinical score of said tissue or organ based on a cellular constituent profile of said tissue or organ, wherein said cellular constituent profile comprises measurements of a plurality of cellular constituent in cells of said tissue or organ after administration of said drug to said animal.

27. The method of claim 26, wherein said composite clinical score of said tissue or organ is determined by a model estimator according to equation

CCS=f(z₁, z₂, . . . , z_n)

where {z₁, z₂, . . . , z_n} are data characterizing said cellular constituent profile.

28. The method of claim 27, wherein said {z₁, z₂, . . . , z_n} are data in a feature space.

29. The method of claim 28, wherein said {z₁, z₂, . . . , z_n} are obtained by transforming said cellular constituent profile using a wavelet transformation of a suitable level.

30. The method of claim 29, wherein said wavelet transformation is a transformation using Daubechies wavelets of a suitable level.

31. The method of claim 26, wherein said model estimator is a neural network model.

32. The method of claim 26, wherein said composite clinical score is a combination of a plurality of k clinical measures of said tissue or organ of said animal.

33. The method of claim 32, wherein each of said plurality of k clinical measures is a converted clinical measure represented as deviations from the respective normal value.

34. The method of claim 33, wherein each of said converted clinical measures is calculated according to the equation

35. The method of claim 34, wherein each of said plurality of k clinical measures is sigmoidal transformed according to the equation

36. The method of claim 35, wherein said composite clinical score is calculated according to the equation

37. The method of claim 35, wherein said organ is liver and said composite clinical score is constructed using a plurality of clinical measures selected from the group consisting of the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), the serum level of cholesterol (Chol), the serum level of gamma-glutamyltranspeptidase (GGT), the serum level of albumin, the serum level of globulins, and the prothrombin time.

38. The method of claim 37, wherein said composite clinical score is constructed using the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), and the serum level of cholesterol (Chol).

39. The method of claim 38, wherein said serum level of alanine aminotransferase (ALT) is sigmoidal transformed with c of 3, and said serum level of aspartate aminotransferase (AST), said serum level of alkaline phosphatase (ALP), said serum level of total bilirubin (Tbil), and said serum level of cholesterol (Chol) are each sigmodal transformed with c of 1.

40. The method of claim 39, wherein said composite clinical score is a hepatotoxicity score HS calculated according to the equation

41. The method of any one of claims 37-40, further comprising classifying said drug according to a predetermined threshold of said composite clinical score, wherein said drug is classified as causing liver damage if said composite clinical score is greater than said predetermined threshold.

79. The method of any one of claims 37-40 and 53-56, wherein said plurality of cellular constituents comprises gene products corresponding to genes or ESTs listed in Table II.

80. The method of claim 79, further comprising measuring said gene products.

42. A method for evaluating the efficacy of a drug in treating a disease or disorder in a tissue or organ in an animal, comprising

(a) determining a composite clinical score of said tissue or organ based on a first cellular constituent profile of said tissue or organ, wherein said first cellular constituent profile comprises measurements of a plurality of cellular constituents in cells of said tissue or organ after administration of said drug to said animal;

and

(b) comparing said composite clinical score determined in step (a) to (b1) standard values of said composite clinical score indicating condition of said tissue or organ;

or (b2) a composite clinical score determined based on a second cellular constituent profile of said tissue or organ, wherein said second cellular constituent profile comprises measurements of said plurality of cellular constituents in cells of said tissue or organ before administration of said drug to said animal;

thereby evaluating the efficacy of said drug in treating said disease.

43. The method of claim 42, wherein said composite clinical score of said tissue or organ is determined by a model estimator according to equation

CCS=f(z₁, z₂, . . . , z_n)

where {z₁, z₂, . . . , z_n} are data characterizing said cellular constituent profile.

44. The method of claim 43, wherein said {z₁, z₂, . . . , z_n} are data in a feature space.

45. The method of claim 44, wherein said {z₁, z₂, . . . , z_n} are obtained by transforming said cellular constituent profile using a wavelet transformation of a suitable level.

46. The method of claim 45, wherein said wavelet transformation is a transformation using Daubechies wavelets of a suitable level.

47. The method of claim 42, wherein said model estimator is a neural network model.

48. The method of claim 42, wherein said composite clinical score is a combination of a plurality of k clinical measures of said tissue or organ of said animal.

49. The method of claim 48, wherein each of said plurality of k clinical measures is a converted clinical measure represented as deviations from the respective normal value.

50. The method of claim 49, wherein each of said converted clinical measures is calculated according to the equation

51. The method of claim 50, wherein each of said plurality of k clinical measures is sigmoidal transformed according to the equation

52. The method of claim 51, wherein said composite clinical score is calculated according to the equation

53. The method of claim 51, wherein said organ is liver and said composite clinical score is constructed using a plurality of clinical measures selected from the group consisting of the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), the serum level of cholesterol (Chol), the serum level of gamma-glutamyltranspeptidase (GGT), albumin, the serum level of globulins, and the prothrombin time.

54. The method of claim 53, wherein said composite clinical score is constructed using the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), and the serum level of cholesterol (Chol).

55. The method of claim 54, wherein said serum level of alanine aminotransferase (ALT) is sigmoidal transformed with c of 3, and said serum level of aspartate aminotransferase (AST), said serum level of alkaline phosphatase (ALP), said serum level of total bilirubin (Tbil), and said serum level of cholesterol (Chol) are each sigmodal transformed with c of 1.

56. The method of claim 55, wherein said composite clinical score is a hepatotoxicity score HS calculated according to the equation

57. A method for determining a model estimator for characterizing a condition of a tissue or organ in an animal, comprising using a plurality of cellular constituent profiles, each comprising measurements of a plurality of cellular constituents, to train a model estimator, said model estimator outputting a composite clinical score using said measurements of said plurality of cellular constituents in a cellular constituent profile, wherein each of said profiles is obtained from said tissue or organ under a different given condition, and wherein each of said profiles has an associated composite clinical score, said composite clinical score being generated using a plurality of clinical measures of said tissue or organ of said animal.

58. The method of claim 57, further comprising before said using step, a step of selecting said plurality of cellular constituent profiles.

59. The method of claim 57, further comprising measuring said plurality of profiles of cellular constituents.

60. The method of claim 57, wherein said model estimator is described by equation

CCS=f(z₁, z₂, . . . , z_n)

where {z₁, z₂ . . . , z_n} are data characterizing said cellular constituent profile.

61. The method of claim 60, wherein said {z₁, z₂, . . . , z_n} are data in a feature space.

62. The method of claim 61, wherein said {z₁, z₂, . . . , z_n} are obtained by transforming said cellular constituent profile using a wavelet transformation of a suitable level.

63. The method of claim 62, wherein said wavelet transformation is a transformation using Daubechies wavelets of a suitable level.

64. The method of claim 57, wherein said model estimator is a neural network model.

65. The method of claim 57, wherein said composite clinical score is determined based on a plurality of k clinical measures of said tissue or organ of said animal.

66. The method of claim 65, wherein each of said plurality of k clinical measures is a converted clinical measure represented as deviations from the respective normal value.

67. The method of claim 66, wherein each of said converted clinical measures is calculated according to the equation

68. The method of claim 67, wherein each of said plurality of k clinical measures is sigmoidal transformed according to the equation

69. The method of claim 68, wherein said composite clinical score is calculated according to the equation

70. The method of claim 68, wherein said organ is liver and said composite clinical score is constructed using a plurality of clinical measures selected from the group consisting of the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), the serum level of cholesterol (Chol), the serum level of gamma-glutamyltranspeptidase (GGT), the serum level of albumin, the serum level of globulins, and the prothrombin time.

71. The method of claim 70, wherein said composite clinical score is constructed using the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), and the serum level of cholesterol (Chol).

72. The method of claim 71, wherein said serum level of alanine aminotransferase (ALT) is sigmoidal transformed with c of 3, and said serum level of aspartate aminotransferase (AST), said serum level of alkaline phosphatase (ALP), said serum level of total bilirubin (Tbil), and said serum level of cholesterol (Chol) are each sigmodal transformed with c of 1.

73. The method of claim 72, wherein said composite clinical score is a hepatotoxicity score HS calculated according to the equation

74. The method of any one of claims 57-73, wherein said condition results from a perturbation to said animal, and said model estimator is used for characterizing an effect of said perturbation on said tissue or organ.

75. The method of claim 74, wherein said perturbation is administration of a drug to said animal, and said effect is a toxicity of said drug.

76. The method of claim 57, wherein said plurality of cellular consituent profiles consists of at least 100 profiles.

77. The method of claims 76, wherein said plurality of cellular consituent profiles consists of at least 1,000 profiles.

78. The method of claims 77, wherein said plurality of cellular consituent profiles consists of at least 10,000 profiles.

81. The method of any one of claims 70-73, wherein said plurality of cellular constituents comprises gene products corresponding to genes or ESTs listed in Table II.

82. The method of claim 81, further comprising measuring said gene products.

83. The method of claim 82, wherein said condition results from a perturbation to said animal, and said model estimator is used for characterizing an effect of said perturbation on said tissue or organ.

84. The method of claim 83, further comprising measuring said gene products.

85. A method of determining hepatotoxicity of a compound at a given dosage in an animal, comprising

(a) contacting hepatocytic cells of said animal with said compound at said dosage;

(b) measuring a cellular constituent profile, wherein said cellular constituent profile comprises measurements of a plurality of cellular constituents in said hepatocytic cells;

(c) determining a composite clinical score of said hepatocytic cells based on said cellular constituent profile; and

(d) determining said compound as having hepatotoxicity if said composite clinical score is above a threshold value.

86. The method of claim 85, wherein said composite clinical score of said tissue or organ is determined by a model estimator according to equation

CCS=f(z₁, z₂, . . . , z_n)

where {Z₁, z₂, . . . , z_n} are data characterizing said cellular constituent profile.

87. The method of claim 86, wherein said {z₁, z₂, . . . , z_n} are data in a feature space.

88. The method of claim 87, wherein said {z₁, z₂, . . . , z_n} are obtained by transforming said cellular constituent profile using a wavelet transformation of a suitable level.

89. The method of claim 88, wherein said wavelet transformation is a transformation using Daubechies wavelets of a suitable level.

90. The method of claim 85, wherein said model estimator is a neural network model.

91. The method of claim 85, wherein said composite clinical score is a combination of a plurality of k clinical measures of said hepatocytic cells of said animal.

92. The method of claim 91, wherein each of said plurality of k clinical measures is a converted clinical measure represented as deviations from the respective normal value.

93. The method of claim 92, wherein each of said converted clinical measures is calculated according to the equation

94. The method of claim 93, wherein each of said plurality of k clinical measures is sigmoidal transformed according to the equation

95. The method of claim 94, wherein said composite clinical score is calculated according to the equation

96. The method of claim 95, wherein said organ is liver and said composite clinical score is constructed using a plurality of clinical measures selected from the group consisting of the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), the serum level of cholesterol (Chol), the serum level of gamma-glutamyltranspeptidase (GGT), the serum level of albumin, the serum level of globulins, and the prothrombin time.

97. The method of claim 96, wherein said composite clinical score is constructed using the serum level of alanine aminotransferase (ALT), the serum level of aspartate aminotransferase (AST), the serum level of alkaline phosphatase (ALP), the serum level of total bilirubin (Tbil), and the serum level of cholesterol (Chol).

98. The method of claim 97, wherein said serum level of alanine aminotransferase (ALT) is sigmoidal transformed with c of 3, and said serum level of aspartate aminotransferase (AST), said serum level of alkaline phosphatase (ALP), said serum level of total bilirubin (Tbil), and said serum level of cholesterol (Chol) are each sigmodal transformed with c of 1.

99. The method of claim 98, wherein said composite clinical score is a hepatotoxicity score HS calculated according to the equation

100. The method of any one of claims 96-99, further comprising classifying said drug according to a predetermined threshold of said composite clinical score, wherein said drug is classified as causing liver damage if said composite clinical score is greater than said predetermined threshold.