Methods for using co-regulated genesets to enhance detection and classification of gene expression patterns

• US 7,599,799 B2
• Filed: 10/18/2002
• Issued: 10/06/2009
• Est. Priority Date: 10/27/1998
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Stoughton, Roland, He, Yudong, Friend, Stephen H.
• Primary Examiner(s): Clow; Lori A
• Application Number:
  US10/273,489
• Publication Number:
  US 20090204333A1
• Time in Patent Office: 2,545 Days
• Field of Search: 702/19, 702/20, 435/6
• US Class Current: 702/19
• CPC Class Codes: C12Q 1/6809 Methods for determination or identification of nucleic acids involving differential detection
• CPC Class Codes: C12Q 1/6834 Enzymatic or biochemical coupling of nucleic acids to a solid phase
• CPC Class Codes: G01N 33/6803 General methods of protein analysis not limited to specific proteins or families of proteins
• 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 25/30 Microarray design
• 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

Abstract

The present invention provides methods for enhanced detection of biological response patterns. In one embodiment of the invention, genes are grouped into basis genesets according to the co-regulation of their expression. Expression of individual genes within a geneset is indicated with a single gene expression value for the geneset by a projection process. The expression values of genesets, rather than the expression of individual genes, are then used as the basis for comparison and detection of biological response with greatly enhanced sensitivity. In another embodiment of the invention, biological responses are grouped according to the similarity of their biological profile.

The methods of the invention have many useful applications, particularly in the fields of drug development and discovery. For example, the methods of the invention may be used to compare biological responses with greatly enhanced sensitivity. The biological responses that may be compared according to these methods include responses to single perturbations, such as a biological response to a mutation or temperature change, as well as graded perturbations such as titration with a particular drug. The methods are also useful to identify cellular constituents, particularly genes, associated with a particular type of biological response. Further, the methods may also be used to identify perturbations, such as novel drugs or mutations, which effect one or more particular genesets. The methods may still further be used to remove experimental artifacts in biological response data.

Claims

1. A method of determining a disease profile that best matches a patient profile, comprising:

(a) comparing, on a suitably programmed computer, a patient profile with reference profiles in a reference database to determine a measure of similarity between said patient profile and each said reference profiles, wherein said patient profile is obtained by projecting a first profile comprising measurements of a plurality of cellular constituents in a cell sample from said patient onto co-varying basis cellular constituent sets, and wherein said reference profiles are obtained by respectively projecting disease profiles each comprising measurements of said plurality of cellular constituents in a disease cell sample onto said co-varying basis cellular constituent sets, said co-varying basis cellular constituent sets being determined based upon co-variation of measurements of cellular constituents, under a plurality of different perturbations;

(b) identifying, on a suitably programmed computer, a reference profile in said reference database that best matches said patient profile based on a maximum similarity among the measures of similarity determined in step (a); and

(c) outputting to a user interface device, a computer readable storage medium, or a local or remote computer system;

or displaying, said maximum similarity or the disease of said disease cell sample of the reference profile in said reference database that best matches said patient profile.

2. The method of claim 1, wherein said co-varying basis cellular constituent sets are determined by cluster analysis of said measurements of cellular constituents, under said plurality of different perturbations.

3. The method of claim 1, wherein said plurality of cellular constituents comprises more than 10 different cellular constituents.

4. The method of claim 3, wherein said plurality of cellular constituents comprises more than 100 different cellular constituents.

5. The method of claim 4, wherein said plurality of cellular constituents comprises more than 1000 different cellular constituents.

6. The method of claim 1, wherein said plurality of different perturbations comprises more than 10 different perturbations.

7. The method of claim 6, wherein said plurality of different perturbations comprises more than 50 different perturbations.

8. The method of claim 7, wherein said plurality of different perturbations comprises more than 100 different perturbations.

9. The method of any one of claims 1-8, wherein said cellular constituents are levels of gene transcripts.

10. The method of any one of claims 1-8, wherein said disease is a cancer.

11. The method of any one of claims 1-9, further comprising projecting said first profile onto said co-varying basis cellular constituent sets to obtain said patient profile.

12. The method of claim 11, wherein said cellular constituents are levels of gene transcripts.

13. The method of any one of claims 1-8, further comprising projecting said disease profiles onto said co-varying basis cellular constituent sets to obtain said reference profiles.

14. The method of any one of claims 1-8, further comprising

(i) projecting said first profile onto said co-varying basis cellular constituent sets to obtain said patient profile;

and

(ii) projecting said disease profiles onto said co-varying basis cellular constituent sets to obtain said reference profiles.

15. The method of claim 1, wherein said measure of similarity is determined according to the equation:

S_ij=S(P_i, P_j)=P_i·

P_j/(|P_i||P_j|).