Cluster analysis of genetic microarray images

• US 7,031,844 B2
• Filed: 03/18/2002
• Issued: 04/18/2006
• Est. Priority Date: 03/18/2002
• Status: Expired due to Fees

• Current Assignee: Board of Regents of the University of Nebraska
• Original Assignee: Board of Regents of the University of Nebraska
• Inventors: Bozinov, Daniel, Rahnenführer, Jörg
• Primary Examiner(s): Zeman, Mary K.
• Application Number:
  US10/100,262
• Publication Number:
  US 20030175720A1
• Time in Patent Office: 1,492 Days
• Field of Search: 702 19- 29, 707/102, 703/11, 435/6
• US Class Current: 702/19
• CPC Class Codes: C12Q 1/6837 using probe arrays or probe chips C12Q1/6874 takes precedence
• CPC Class Codes: C12Q 2537/161 A competitive reaction step
• CPC Class Codes: C12Q 2565/102 Multiple non-interacting labels
• CPC Class Codes: C40B 30/04 by measuring the ability to specifically bind a target molecule, e.g. antibody-antigen binding, receptor-ligand binding
• CPC Class Codes: G01N 33/5308 for analytes not provided for elsewhere, e.g. nucleic acids, uric acid, worms, mites
• CPC Class Codes: G01N 33/582 with fluorescent label
• CPC Class Codes: G01N 33/6845 Methods of identifying protein-protein interactions in protein mixtures
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06T 2207/30072 Microarray; Biochip, DNA array; Well plate
• CPC Class Codes: G06T 7/0012 Biomedical image inspection
• CPC Class Codes: G06T 7/90 Determination of colour characteristics
• CPC Class Codes: G16B 25/00 ICT specially adapted for hybridisation; ICT specially adapted for gene or protein expression
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A method for determining relative incidence of a binding substance within two biological samples is provided. The two samples are labeled with luminescent materials having different chromatic properties. An image of the luminescent materials upon a binding site of a microarray is analyzed as two clusters of data points scattered about respective representative pairs of chromatic intensity values. The relative incidence of the binding substance is determined as a ratio of differences between corresponding indices of the representative pairs.

Claims

1. A method of determining the relative abundance of genetic material present in a first sample and a second sample, comprising the steps of:

labeling the first sample with a first detectable label adapted to bind with genetic material therein;

labeling the second sample with a second detectable label adapted to bind with genetic material therein;

providing a substrate with a binding area thereon for hybridizing with a genetic sequence;

contacting the substrate comprising said genetic sequence with the first and second samples under conditions suitable to effect hybridization;

obtaining a digital image of the binding area under conditions causing said labels to manifest an image intensity dependent on the abundance of genetic material bound to the binding area, if any, thus providing an array of pixel values comprising respective intensity values corresponding the first and second labels;

assigning each of said pixel values to a respective one of a foreground cluster and a background cluster;

determining a representative statistical vector of pixel values within each of said foreground cluster and said background cluster; and

computing the relative abundance of genetic material within the first and second samples corresponding to the binding area in proportion to a ratio of differences between the respective coordinates of the representative statistical vectors.

2. The method of claim 1 wherein said representative statistical vectors are a pair of medoid vectors for the respective clusters, and wherein said respective clusters comprise a foreground cluster and a background cluster.

3. The method of claim 2 wherein said determining step is one of a Partition Around Medoids analysis step and a K-Means analysis step.

4. The method of claim 3, further comprising the steps of:

generating a bivalent mask of the respective background and foreground areas of the binding area based on the outcome of said clustering procedure; and

re-applying said clustering procedure to pixels located within a determined foreground cluster, prior to providing said measure of abundance.

5. The method of claim 3, further comprising the steps of:

calculating the spatial dispersion of image pixels corresponding to said determined foreground cluster;

calculating the spatial dispersion of image pixels corresponding to said determined background cluster; and

reversing the determination of foreground and background clusters if the spatial dispersion of said determined foreground cluster exceeds the spatial dispersion of said background cluster.

6. The method of claim 3, further comprising the steps of:

generating a bivalent mask of the respective background and foreground areas of the binding area based on the outcome of said clustering procedure; and

generating a display of said bivalent mask for visual comparison with the image of the binding area.