System and method for generating, analyzing and storing normalized expression datasets from raw expression datasets derived from microarray includes nucleic acid probe sequences

• US 6,408,308 B1
• Filed: 06/29/1998
• Issued: 06/18/2002
• Est. Priority Date: 01/29/1998
• Status: Expired due to Fees

• Current Assignee: Incyte Pharmaceuticals, Inc.
• Original Assignee: Incyte Pharmaceuticals, Inc.
• Inventors: Cocks, Benjamin G., Kerr, James R., Rao, Srikar D., Engler, Timothy M., Lassagne, Steven M., Seilhamer, Jeffrey J., Maslyn, Timothy J., Cheng, Rachel J.
• Primary Examiner(s): Breene, John
• Assistant Examiner(s): CHANNAVAJJALA, SRIRAMA T
• Application Number:
  US09/106,607
• Time in Patent Office: 1,450 Days
• Field of Search: 702/19-21, 702/27, 707/1-6, 707/100-104, 435/3-6, 435/303, 435/71-- 7, 435/287.2, 436/278-279, 706/924, 703/11-12, 382/128-129, 536/24.5
• US Class Current: 707/776
• CPC Class Codes: G16B 25/00 ICT specially adapted for hybridisation; ICT specially adapted for gene or protein expression
• CPC Class Codes: G16B 25/10 Gene or protein expression profiling; Expression-ratio estimation or normalisation
• CPC Class Codes: G16B 45/00 ICT specially adapted for bioinformatics-related data visualisation, e.g. displaying of maps or networks
• CPC Class Codes: G16B 50/00 ICT programming tools or database systems specially adapted for bioinformatics
• CPC Class Codes: Y10S 707/941 Human sciences
• CPC Class Codes: Y10S 707/99945 Object-oriented database structure processing

Abstract

A biomolecular expression information processing system has procedures and tables that store abundance datasets and hybridization data. The tables also store information identifying a microarray technology type for each hybridization and microarray design information for each microarray technology type. The microarray design information includes technology data that specifies global characteristics of each microarray, and array element data that specifies characteristics of array elements in each microarray instance of the microarray technology type. The procedures process the abundance datasets in accordance with the microarray design information associated with each such abundance dataset. The system stores technology data for multiple distinct microarray technology types and stores array element data for multiple microarray designs of a single technology type.

Claims

1. A method for storing normalized expression datasets from raw expression datasets, comprising:

storing one or more raw expression datasets derived from at least one microarray, wherein the at least one microarray includes nucleic acid probe sequences, wherein each raw expression dataset comprises a plurality of expression intensities of the nucleic acid probe sequences on the at least one microarray;

executing at least one normalization procedure that processes the raw expression datasets in accordance with the microarray design information associated with each raw expression dataset;

generating one or more normalized datasets from the raw expression datasets using one or more normalization procedures;

storing the normalized datasets based on the raw expression datasets and the procedures; and

storing procedure identifiers representing the one or more procedures that were used to generate the normalized datasets from the raw expression datasets;

whereby one or more normalized datasets corresponding to a single raw expression dataset are stored.

2. The method of claim 1 further comprising:

selecting one of the normalized datasets as a base expression dataset;

selecting another of the normalized datasets as a comparison dataset; and

generating a set of expression ratio values representing ratios of the expression intensities in the selected and base expression datasets.

3. The method of claim 1 wherein generating one or more normalized datasets comprises:

generating normalized data from a first raw expression dataset using a first normalization procedure,

generating normalized data from the first raw expression dataset using a second normalization procedure;

wherein storing one or more normalized datasets includes;

storing a first normalized dataset based on the first raw expression dataset and the first normalization procedure;

storing a second normalized dataset based on the first raw expression dataset and the second normalization procedure; and

wherein storing normalization procedure identifiers stores normalization procedure identifiers representing the first and second normalization procedures that were used to generate the first and second normalized datasets from the first raw expression dataset;

whereby two distinct normalized datasets corresponding to a single expression dataset are stored.

4. The method of claim 3 further comprising:

storing the first and second normalization procedures; and

generating new normalized data from the first one of the raw expression datasets using a new normalization procedure;

storing a new normalized dataset based on the new normalized data and the new normalization procedure; and

storing a new normalization procedure identifier representing the new normalization procedure; and

storing the new normalization procedure.

5. The method of claim 3 wherein the first and second normalization procedures include first and second summarization methods, respectively, and a first normalization determination method, and further comprising:

storing the first and second summarization methods;

storing the first normalization determination method;

generating new summarization data from the first one of the stored raw expression datasets using a new summarization method;

storing the new summarization dataset based on the new summarization data and the new summarization method;

storing a summarization method identifier for the new summarization method;

storing the new summarization method;

generating new normalized data from the new summarization dataset using the first normalization determination method; and

storing a new normalized dataset based on the new normalized data and first normalization determination method.

6. The method of claim 3 wherein the first and second procedures include first and second normalization determination methods, respectively, and

further comprising:

storing a first summarization dataset based on the first one of the historical raw expression datasets and the first normalization procedure;

storing the first and second normalization determination methods;

generating new normalized data from the first summarization dataset using a new normalization determination method;

storing a new normalized dataset based on the new normalized data and new normalization determination method;

storing a normalization determination method identifier for the new normalization determination method; and

storing the new normalization determination method.

7. The method of claim 1, further comprising storing the one or more procedures.

12. The biomolecular expression information processing system of claim 7 further comprising:

a dataset comparison procedure that

selects a first normalized dataset as a base expression dataset;

selects a second normalized dataset as a comparison dataset; and

generates a set of expression ratio values representing ratios of expression intensities in the comparison and base expression datasets.

8. A biomolecular expression information processing system, comprising:

one or more tables for storing;

normalized datasets of expression intensities from nucleic acid probe sequences;

hybridization data that identifies, for each of at least a subset of the normalized datasets, a hybridization from which the normalized dataset was generated, the hybridization comprising a sample and microarray to which the sample was applied;

information identifying, for each hybridization, a microarray technology type;

microarray design information, for each microarray technology type, including

technology data specifying global characteristics of each microarray instance of the microarray technology type;

array element data specifying characteristics of array elements of nucleic acid probe sequences in each microarray instance of the microarray technology type;

procedures, executed by the system, that process the normalized datasets in accordance with the microarray design information associated with each such normalized dataset;

wherein the system is capable of storing technology data for multiple distinct microarray technology types and is capable of storing array element data for multiple microarray designs of a single technology type.

9. The biomolecular expression information processing system of claim 8 wherein one microarray technology type generates a first raw expression dataset from a single hybridization, wherein the first raw expression dataset comprises a plurality of expression intensities of nucleic acid probe sequences.

10. The biomolecular expression information processing system of claim 8 wherein one microarray technology generates at least two raw expression datasets from a single hybridization, wherein the at least two raw expression datasets each comprise a plurality of expression intensities of nucleic acid probe sequences.

11. The biomolecular expression information processing system of claim 8 wherein one microarray technology type generates a first raw expression dataset for a single hybridization, and another microarray technology type generates at least two raw expression datasets from a single hybridization, wherein the first dataset and the at least two datasets each comprise a plurality of expression intensities of nucleic acid probe sequences.

13. A method of processing biomolecular expression data, comprising:

selecting a set of expression datasets, wherein each expression dataset data comprises a plurality nucleic acid probe sequence elements;

selecting a subset of the elements in each of the selected expression datasets in accordance with specified selection criteria;

storing in one or more pseudoarrays the selected subset of elements for selected expression datasets; and

using the one or more pseudoarrays, comparing corresponding elements of the different selected expression datasets to identify expression differences between the selected expression datasets.

14. The method of claim 13 wherein at least two of the selected expression datasets are associated with different microarray technologies.

15. The method of claim 13 wherein at least one expression dataset is from a microarray.

16. The method of claim 13 wherein at least one expression dataset is from an expression database.

17. The method of claim 13 wherein said step of selecting expression datasets includes the step of selecting a protein function, displaying identification information for expression datasets with elements having the selected protein function, selecting at least one expression dataset using the displayed identification information.

18. The method of claim 13 wherein at least a subset of the elements has an intensity, and said step of selecting a subset of the elements selects those elements having an intensity exceeding a predetermined threshold.

19. The method of claim 13 wherein at least a subset of the elements has an intensity, and said step of selecting a subset of the elements selects those elements having an intensity outside a predetermined range.

20. The method of claim 13 wherein at least a subset of the elements has an intensity, and said step of selecting a subset of the elements selects those elements having an intensity within a predetermined range.

21. The method of claim 13 wherein at least a subset of the elements has a hit description, and said step of selecting a subset of the elements selects those elements having a particular hit description.

22. The method of claim 13 wherein at least a subset of the elements has a transcript identifier, and said step of selecting a subset of the elements selects those elements having a particular transcript identifier.

23. The method of claim 13 wherein the pseudoarrays are compared by displaying the set of selected elements of each pseudoarray wherein each element is displayed as a colored spot with a color-coded intensity.

24. The method of claim 13 wherein the pseudoarrays are compared by displaying the set of identified elements in a table showing a numerical value for the intensity.

25. The method of claim 13 wherein selected expression datasets comprise a comparison expression dataset and a base expression dataset, and a set of expression ratio values are generated representing ratios of expression intensities in the comparison and base expression datasets.

26. The method of claim 13 wherein selected expression datasets comprise a comparison expression dataset and a base expression dataset, and a set of absolute expression values are generated representing a difference of expression intensities between the comparison and base expression datasets.

27. A computer-readable medium for storing instructions for a computer to store normalized expression datasets from raw expression datasets, the instructions comprising:

generating at least one normalized dataset from at least one raw expression dataset using at least one normalization, wherein the raw expression dataset comprises at least one expression intensity of nucleic acid probe sequences from at least one microarray; and

storing the at least one normalized dataset based on the at least one raw expression dataset and the at least one procedure.

28. The computer-readable medium of claim 27, the instructions further comprising storing procedure identifiers representing the at least one procedure.

29. The computer-readable medium of claim 27, the instructions further comprising storing the at least one raw expression dataset.

30. The computer-readable medium of claim 27, wherein the at least one normalized dataset includes a first normalized dataset and a second normalized dataset,

the instructions further comprising:

selecting the first normalized dataset as a base expression dataset;

selecting the second normalized dataset as a comparison dataset; and

generating a set of expression ratio values representing ratios of the expression intensities in the comparison and base expression datasets.

31. The computer-readable medium of claim 27, the instructions further comprising storing the at least one procedure.

32. The computer-readable medium of claim 27, the instructions further comprising summarizing the at least one expression intensity using a summarization method.

33. The computer-readable medium of claim 32, the instructions further comprising storing the summarization method.

34. The computer-readable medium of claim 32, the instructions further comprising storing summarization method identifiers representing the summarization method.

35. A computer-readable medium for storing instructions for a computer to process biomolecular expression data, the instructions comprising:

selecting a set of expression datasets, each expression dataset having a plurality of nucleic acid probe sequence elements;

selecting a subset of the elements in each of the selected expression datasets in accordance with specified selection criteria;

storing in one or more pseudoarrays the selected subset of elements for selected expression datasets; and

using the one or more pseudoarrays, comparing corresponding elements of the different selected expression datasets to identify expression differences between the selected expression datasets.

36. The computer-readable medium of claim 35 wherein at least two of the selected expression datasets are associated with different microarray technologies.

37. The computer-readable medium of claim 35 wherein at least one expression dataset is from a microarray.

38. The computer-readable medium of claim 35 wherein at least one expression dataset is from an expression database.

39. The computer-readable medium of claim 35 wherein selecting expression datasets includes selecting a protein function, displaying identification information for expression datasets with elements having the selected protein function, selecting at least one expression dataset using the displayed identification information.

40. The computer-readable medium of claim 35 wherein at least a subset of the elements has an intensity, and selecting a subset of the elements selects those elements having an intensity exceeding a predetermined threshold.

41. The computer-readable medium of claim 35 wherein at least a subset of the elements has an intensity, and selecting a subset of the elements selects those elements having an intensity outside a predetermined range.

42. The computer-readable medium of claim 35 wherein at least a subset of the elements has an intensity, and selecting a subset of the elements selects those elements having an intensity within a predetermined range.

43. The computer-readable medium of claim 35 wherein at least a subset of the elements has a hit description, and selecting a subset of the elements selects those elements having a particular hit description.

44. The computer-readable medium of claim 35 wherein at least a subset of the elements has a transcript identifier, and selecting a subset of the elements selects those elements having a particular transcript identifier.

45. The computer-readable medium of claim 35 wherein the pseudoarrays are compared by displaying the set of selected elements of each pseudoarray wherein each element is displayed as a colored spot with a color-coded intensity.

46. The computer-readable medium of claim 35 wherein the pseudoarrays are compared by displaying the set of identified elements in a table showing a numerical value for the intensity.

47. The computer-readable medium of claim 35 wherein selected expression datasets comprise a comparison expression dataset and a base expression dataset, and a set of expression ratio values are generated representing ratios of expression intensities in the comparison and base expression datasets.

48. The computer-readable medium of claim 35 wherein selected expression datasets comprise a comparison expression dataset and a base expression dataset, and a set of absolute expression values are generated representing a difference of expression intensities between the comparison and base expression datasets.

49. A system to store normalized expression datasets from raw expression datasets, comprising:

means for generating at least one normalized dataset from at least one raw expression dataset using at least one normalization procedure, wherein the raw expression dataset comprises at least one expression intensity of nucleic acid probe sequences from at least one microarray; and

means for storing the at least one normalized dataset based on the at least one raw expression dataset and the at least one procedure.

50. A system for processing biomolecular expression data, comprising:

means for selecting a set of expression datasets, each expression dataset having a plurality of nucleic acid probe sequence elements;

means for selecting a subset of the elements in each of the selected expression datasets in accordance with specified selection criteria;

means for storing in one or more pseudoarrays the selected subset of elements for selected expression datasets; and

means for using the one or more pseudoarrays, comparing corresponding elements of the different selected expression datasets to identify expression differences between the selected expression datasets.