Identifying antigen clusters for monitoring a global state of an immune system

• US 20040014069A1
• Filed: 01/06/2003
• Published: 01/22/2004
• Est. Priority Date: 07/24/2000
• Status: Abandoned Application

• Current Assignee: Yeda Research and Development Co. Ltd. (Weizmann Institute of Science)
• Original Assignee: Yeda Research and Development Co. Ltd. (Weizmann Institute of Science)
• Inventors: Hed, Guy, Cohen, Irun R., Getz, Gad, Domany, Eytan, Quintana, Francisco Javier
• Application Number:
  US10/332,241
• Publication Number:
  US 20040014069A1
• Time in Patent Office: Days
• US Class Current: 435/6
• CPC Class Codes: G01N 33/564 for pre-existing immune complex or autoimmune disease , i.e. systemic lupus erythematosus, rheumatoid arthritis, multiple sclerosis, rheumatoid factors or complement components C1-C9
• CPC Class Codes: G01N 33/6803 General methods of protein analysis not limited to specific proteins or families of proteins
• CPC Class Codes: G16B 20/00 ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
• 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 40/20 Supervised data analysis
• CPC Class Codes: G16B 40/30 Unsupervised data analysis

Abstract

Method, system and an article of manufacture for clustering and thereby identifying predefined antigens reactive with undetermined immunoglobulins of sera derived from patient subjects in need of diagnosis of disease or monitoring of treatment.

Claims

1. A method of classifying into a predefined first situation of at least two distinct situations, a binding pattern of a plurality of undetermined first binding moieties, said plurality of undetermined first binding moieties being derived from a first group of objects being associated with said predefined first situation and from at least one second group of objects being associated with a situation other than said first situation of said at least two distinct situations, to a predefined set of a plurality of potential second binding moieties, the method comprising the steps of:

(a) assaying said plurality of undetermined first binding moieties of said first group of objects for binding to each of said plurality of potential second binding moieties;

(b) assaying said plurality of undetermined first binding moieties of said at least one second group of objects for binding to each of said plurality of potential second binding moieties; and

(c) clustering at least some of said plurality of potential second binding moieties into clusters of second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects, thereby classifying into said predefined first situation said binding pattern of said plurality of undetermined first binding moieties of said objects being associated with said predefined first situation to said predefined set of said plurality of potential second binding moieties.

2. The method of claim 1, wherein said step of clustering at least some of said plurality of potential second binding moieties into clusters of second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects is effected by a supervised classifier.

3. The method of claim 2, wherein said supervised classifier is a neural network algorithm.

4. The method of claim 1, wherein said step of clustering at least some of said plurality of potential second binding moieties into clusters of second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects is effected by a unsupervised classifier.

5. The method of claim 4, wherein said unsupervised classifier is a coupled two way clustering algorithm.

6. The method of claim 1, further comprising the step of:

(d) scanning said second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects and selecting for a subset of said second binding moieties resulting in an optimal sensitivity.

7. The method of claim 1, further comprising the step of:

(d) scanning said second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects and selecting for a subset of said second binding moieties resulting in an optimal specificity.

8. The method of claim 1, further comprising the step of:

(d) scanning said second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects and selecting for a subset of said second binding moieties resulting in an optimal specificity and an optimal sensitivity.

9. The method of claim 1, wherein said first binding moieties are immunoglobulins, whereas said second binding moieties are antigens.

10. The method of claim 1, wherein said first situation is a human disease.

11. The method of claim 10, wherein said human disease is selected from the group consisting of an autoimmune disease, a cancer, an immune deficiency disease, a degenerative disease, a metabolic disease, an infectious disease, a genetic disease, a mental disorder, an organ transplantation, an injury or an intoxication, or any condition involving cytokines or inflamation.

12. The method of claim 1, wherein said first binding moieties and said second binding moieties are each independently selected from the group consisting of nucleic acids, proteins, peptides, carbohydrates, fatty acids, synthetic compounds, tissue extracts, or peptide or expression libraries.

13. A method of classifying a specific object into a situation of at least two distinct situations, the method comprising the steps of:

(a) classifying binding patterns of a plurality of undetermined first binding moieties, said plurality of undetermined first binding moieties being derived from a first group of objects being associated with a predefined first situation of said at least two distinct situation, and from at least one second group of objects being associated with a situation other than said first situation of said at least two distinct situations, to a predefined set of a plurality of potential second binding moieties by clustering at least some of said plurality of potential second binding moieties into a cluster of second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects; and

(b) using said cluster for determining whether said specific object is classifiable into said situation of said at least two distinct situations.

14. The method of claim 13, wherein said step of clustering at least some of said plurality of potential second binding moieties into clusters of second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects is effected by a supervised classifier.

15. The method of claim 14, wherein said supervised classifier is a neural network algorithm.

16. The method of claim 13, wherein said step of clustering at least some of said plurality of potential second binding moieties into clusters of second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects is effected by a unsupervised classifier.

17. The method of claim 16, wherein said unsupervised classifier is a coupled two way clustering algorithm.

18. The method of claim 13, wherein step (a) includes the step of:

(i) scanning said second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects and selecting for a subset of said second binding moieties resulting in an optimal sensitivity.

19. The method of claim 13, wherein step (a) includes the step of:

(i) scanning said second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects and selecting for a subset of said second binding moieties resulting in an optimal specificity.

20. The method of claim 13, wherein step (a) includes the step of:

(i) scanning said second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects and selecting for a subset of said second binding moieties resulting in an optimal specificity and an optimal sensitivity.

21. The method of claim 13, wherein said first binding moieties are immunoglobulins, whereas said second binding moieties are antigens.

22. The method of claim 13, wherein said first situation is a human disease.

23. A method of clustering a subset of antigens of a plurality of antigens, said subset of antigens being reactive with a plurality of antibodies being derived from a plurality of patients having an impaired immune system and suffering from a disease, the method comprising the steps of:

(a) assaying binding of said plurality of antibodies being derived from said plurality of patients with said plurality of antigens;

(b) assaying binding of a plurality of antibodies being derived from a plurality of individuals free of said disease with said plurality of antigens; and

(c) clustering said subset of antigens being reactive with said plurality of antibodies being derived from said plurality of patients having said impaired immune system and suffering from said disease.

24. The method of claim 23, wherein said step of clustering is effected so as to include in said subset of antigens those antigens for which said patients and individuals best decompose into clusters according to a known clinical diagnosis of said patients and individuals.

25. The method of claim 23, wherein said step of clustering is effected by a supervised classifier.

26. The method of claim 25, wherein said supervised classifier is a neural network algorithm.

27. The method of claim 23, wherein said step of clustering is effected by a unsupervised classifier.

28. The method of claim 27, wherein said unsupervised classifier is a coupled two way clustering algorithm.

29. The method of claim 23, wherein said step of clustering is effected so as to result in optimal sensitivity.

30. The method of claim 23, wherein said step of clustering is effected so as to result in optimal specificity.

31. The method of claim 23, wherein said step of clustering is effected so as to result in optimal specificity and optimal sensitivity.

32. The method of claim 23, wherein said step of clustering is effected by:

(i) clustering said antibodies and said antigens and identifying all stable antibody and antigen clusters;

(ii) scanning said antigen clusters, while using reactivity levels of antigens of each antigen cluster as a feature set representing first object sets containing either all of said antibodies or any of said stable antibody clusters;

(iii) scanning said antibody clusters, while using reactivity levels of antibody of each antibody cluster as a feature set representing second object sets containing either all of said antigens or any of said stable antigen clusters;

(iv) tracking all antibody and antigen stable clusters thus generated;

(v) repeating steps (i)-(iv) until no new antibody and antigen stable clusters being generated, thereby obtaining final stable antigens and antibody clusters and pointers identifying how all of said stable antibody and antigen clusters have been generated.

33. The method of claim 23, wherein said disease is selected from the group consisting of a autoimmune disease, a cancer, an immune deficiency disease, a degenerative disease, a metabolic disease, an infectious disease, a genetic disease, a mental disorder, an organ transplantation, an injury or an intoxication, or any condition involving cytokines or inflamation.

34. The method of claim 33, wherein said autoimmune disease is selected from the group consisting of ankylosing spondylitis, uveitis, Goodpasture'"'"'s syndrome, multiple sclerosis, Grave'"'"'s disease, myasthenia gravis, systemic lupus erythematosus, systemic sclerosis, mixed connective tissue disease, dermatitis herpetiformis, celiac disease, ulcerative colitis, Crohn'"'"'s disease, chronic active hepatitis, endometriosis, ulcerative colitis, insulin-dependent diabetes mellitus, psoriasis, pemphingus vulgaris, Hashimoto'"'"'s thyroiditis, rheumatoid arthritis, idiopathic thrombocytopenic purpura, Sjogren'"'"'s syndrome, uveroretinitis, autoimmune hemolytic anemia, vitiligo, primary biliary cirrhosis, inflammatory bowel disease, Bechet'"'"'s disease, auricular chondritis, tympanosclerosis, autoimmune salpingitis, otosclerosis, secretory otitis media, necrotizing otitis media, autoimmune sensorineural hearing loss, Meniere'"'"'s disease and cochlear vasculitis.

35. A method of diagnosing a disease of a subject, the method comprising the steps of:

(a) clustering a subset of antigens of a plurality of antigens, said subset of antigens being reactive with a plurality of antibodies being derived from a plurality of patients having an impaired immune system and suffering from said disease by;

(i) assaying binding of said plurality of antibodies being derived from said plurality of patients with said plurality of antigens;

(ii) assaying binding of a plurality of antibodies being derived from a plurality of individuals free of said disease with said plurality of antigens; and

(iii) clustering said subset of antigens being reactive with said plurality of antibodies being derived from said plurality of patients having said impaired immune system and suffering from said disease; and

(b) associating or deassociating the antibodies of said subject with a cluster resulting from step (a)(iii).

36. The method of claim 35, wherein said step of clustering is effected so as to include in said subset of antigens those antigens for which said patients and individuals best decompose into clusters according to a known clinical diagnosis of said patients and individuals.

37. The method of claim 35, wherein said step of clustering is effected by a supervised classifier.

38. The method of claim 37, wherein said supervised classifier is a neural network algorithm.

39. The method of claim 35, wherein said step of clustering is effected by a unsupervised-classifier.

40. The method of claim 39, wherein said unsupervised classifier is a coupled two way clustering algorithm.

41. The method of claim 35, wherein said step of clustering is effected so as to result in optimal sensitivity.

42. The method of claim 35, wherein said step of clustering is effected so as to result in optimal specificity.

43. The method of claim 35, wherein said step of clustering is effected so as to result in optimal specificity and optimal sensitivity.

44. A system for monitoring the state of the immune system of a subject, the system comprising a data acquisition device and a computation device communicating therewith, said data acquisition device and said computation device being designed, constructed and configured for:

(a) clustering a subset of antigens of a plurality of antigens, said subset of antigens being reactive with a plurality of antibodies being derived from a plurality of patients having an impaired immune system and suffering from said disease by;

(i) assaying binding of said plurality of antibodies being derived from said plurality of patients with said plurality of antigens;

(ii) assaying binding of a plurality of antibodies being derived from a plurality of individuals free of said disease with said plurality of antigens; and

(iii) clustering said subset of antigens being reactive with said plurality of antibodies being derived from said plurality of patients having said impaired immune system and suffering from said disease; and

(b) associating or deassociating serum of said subject with a cluster resulting from step (a)(iii).

45. The method of claim 44, wherein said step of clustering is effected by:

(i) clustering said antibodies and said antigens and identifying all stable antibody and antigen clusters;

(ii) scanning said antigen clusters, while using reactivity levels of antigens of each antigen cluster as a feature set representing first object sets containing either all of said antibodies or any of said stable antibody clusters;

(iii) scanning said antibody clusters, while using reactivity levels of antibodies of each antibody cluster as a feature set representing second object sets containing either all of said antigens or any of said stable antigen clusters;

(iv) tracking all antibody and antigen stable clusters thus generated;

(v) repeating steps (i)-(iv) until no new antibody and antigen stable clusters being generated, thereby obtaining final stable antigen and antibody clusters and pointers identifying how all of said stable antibody and antigen clusters have been generated.

46. The method of claim 44, wherein said disease is selected from the group consisting of a autoimmune disease, a cancer, an immune deficiency disease, a degenerative disease, a metabolic disease, an infectious disease, a genetic disease, a mental disorder, an organ transplantation, an injury or an intoxication, or any condition involving cytokines or inflamation.

47. The method of claim 46, wherein said autoimmune disease is selected from the group consisting of ankylosing spondylitis, uveitis, Goodpasture'"'"'s syndrome, multiple sclerosis, Grave'"'"'s disease, myasthenia gravis, systemic lupus erythematosus, systemic sclerosis, mixed connective tissue disease, dermatitis herpetiformis, celiac disease, ulcerative colitis, Crohn'"'"'s disease, chronic active hepatitis, endometriosis, ulcerative colitis, insulin-dependent diabetes mellitus, psoriasis, pemphingus vulgaris, Hashimoto'"'"'s thyroiditis, rheumatoid arthritis, idiopathic thrombocytopenic purpura, Sjogren'"'"'s syndrome, uveroretinitis, autoimmune hemolytic anemia, vitiligo, primary biliary cirrhosis, inflammatory bowel disease, Bechet'"'"'s disease, auricular chondritis, tympanosclerosis, autoimmune salpingitis, otosclerosis, secretory otitis media, necrotizing otitis media, autoimmune sensorineural hearing loss, Meniere'"'"'s disease and cochlear vasculitis.

48. An article of manufacture comprising a surface and antigens being arranged on said surface, each in an independent addressable location, said antigens including a subset of antigens being selected by a method of clustering said subset of antigens of a plurality of antigens, said subset of antigens being reactive with a plurality of antibodies being derived from a plurality of patients having an impaired immune system and suffering from a disease, the method being effected by:

(a) assaying binding of said plurality of antibodies being derived from said plurality of patients with said plurality of antigens;

(b) assaying binding of a plurality of antibodies being derived from a plurality of individuals free of said disease with said plurality of antigens; and

(c) clustering said subset of antigens being reactive with said plurality of antibodies being derived from said plurality of patients having said impaired immune system and suffering from said disease.

49. An article of manufacture comprising a surface and antigens being arranged on said surface, each in an independent addressable location, said antigens including a plurality of subsets of antigens, each of said plurality of subsets of antigens being selected by a method of clustering a subset of antigens of a plurality of antigens, said subset of antigens being reactive with a plurality of antibodies being derived from a plurality of patients having an impaired immune system and suffering from a specific disease, the method being effected by:

(a) assaying binding of said plurality of antibodies being derived from said plurality of patients with said plurality of antigens;

(b) assaying binding of a plurality of antibodies being derived from a plurality of individuals free of said disease with said plurality of antigens; and

(c) clustering said subset of antigens being reactive with said plurality of antibodies being derived from said plurality of patients having said impaired immune system and suffering from said disease.

50. A system for classifying into a predefined first situation of at least two distinct situations, a binding pattern of a plurality of undetermined first binding moieties, said plurality of undetermined first binding moieties being derived from a first group of objects being associated with said predefined first situation and from at least one second group of objects being associated with a situation other than said first situation of said at least two distinct situations, to a predefined set of a plurality of potential second binding moieties, the system comprising a data acquisition device and a computation device communicating therewith, said data acquisition device and said computation device being designed, constructed and configured for:

(a) assaying said plurality of undetermined first binding moieties of said first group of objects for binding to each of said plurality of potential second binding moieties;

(b) assaying said plurality of undetermined first binding moieties of said at least one second group of objects for binding to each of said plurality of potential second binding moieties; and

(c) clustering at least some of said plurality of potential second binding moieties into clusters of second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects, thereby classifying into said predefined first situation said binding pattern of said plurality of undetermined first binding moieties of said objects being associated with said predefined first situation to said predefined set of said plurality of potential second binding moieties.

51. A system for classifying a specific object into a situation of at least two distinct situations, the system comprising a data acquisition device and a computation device communicating therewith, said data acquisition device and said computation device being designed, constructed and configured for:

(a) classifying binding patterns of a plurality of undetermined first binding moieties, said plurality of undetermined first binding moieties being derived from a first group of objects being associated with a predefined first situation of said at least two distinct situation, and from at least one second group of objects being associated with a situation other than said first situation of said at least two distinct situations, to a predefined set of a plurality of potential second binding moieties by clustering at least some of said plurality of potential second binding moieties into a cluster of second binding moieties which bind first binding moieties of said undetermined first binding moieties from said first group of objects; and

(b) using said cluster for determining whether said specific object is classifiable into said situation of said at least two distinct situations.

52. A system for clustering a subset of antigens of a plurality of antigens, said subset of antigens being reactive with a plurality of antibodies being derived from a plurality of patients having an impaired immune system and suffering from a disease, the system comprising a data acquisition device and a computation device communicating therewith, said data acquisition device and said computation device being designed, constructed and configured for:

(a) assaying binding of said plurality of antibodies being derived from said plurality of patients with said plurality of antigens;

(b) assaying binding of a plurality of antibodies being derived from a plurality of individuals free of said disease with said plurality of antigens; and

(c) clustering said subset of antigens being reactive with said plurality of antibodies being derived from said plurality of patients having said impaired immune system and suffering from said disease.