Microarray Spot Reliability Determination Using Median Intensity Ratios

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for evaluating the uniformity of spots in microarray experiments are inadequate, leading to excessive detection of defective spots due to non-biological factors like dust or uneven distribution, which results in the elimination of usable data.

Innovation Solution

A determination method that calculates the ratio or difference between the median detection intensity of a pixel group and the median intensity excluding a certain top or bottom proportion of pixels, using expressions such as |X - Xt| / X or |X - Xb| / X, to assess the reliability of selective binding, with a reference value determined by S = C + Z / X, where S is the reference value, C is a constant, Z is an offset value, and X is the median intensity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the coefficient of variation (CV) method is used to evaluate spot uniformity, then the detection of defective spots becomes more sensitive, but usable data is excessively eliminated due to non-biological factors like dust or uneven distribution

Engineering Contradiction:
Improvedefective spot detection accuracyVSAvoidloss of usable data
Core Design Contradiction:
ReliabilityVSLoss of substance

Solution Approach 1:

The patent segments the pixel group into multiple regions (central region and peripheral regions) and calculates median values for each segment. By comparing the central region median with peripheral region medians, the method can identify localized defects (like dust) without eliminating the entire spot data, thus reducing excessive data loss while maintaining defect detection capability

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary evaluation metric based on median value comparisons between different pixel group regions. This intermediary approach provides a more nuanced assessment than the overall CV method, allowing differentiation between acceptable variations and actual defects, thereby preventing unnecessary elimination of usable data

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If the median of pixel group is used as representative value, then the impact of outliers is reduced, but uniformity within single spot cannot be evaluated

Engineering Contradiction:
Improvesignal intensity measurement accuracyVSAvoidspot uniformity quality
Core Design Contradiction:
Measurement precisionVSManufacturing precision

Solution Approach 1:

The patent divides the pixel group into a central region (first pixel group) and peripheral regions (second pixel groups). By calculating and comparing median values of these segmented regions, the method simultaneously achieves robust outlier resistance (through median calculation) and spot uniformity evaluation (through regional comparison)

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent transitions from a single-dimensional median evaluation to a multi-dimensional evaluation by comparing medians across different spatial regions (central vs. peripheral). This dimensional expansion enables assessment of both measurement accuracy and spot uniformity simultaneously

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

3Reliability

If existing uniformity evaluation methods are applied, then spot-to-spot variation can be assessed, but non-biological factors cause false positive defect detection

Engineering Contradiction:
Improveuniformity evaluation capabilityVSAvoidnon-biological interference
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent applies local quality analysis by evaluating different regions of the spot differently. The central region median is compared with peripheral region medians to identify localized anomalies. This local evaluation approach distinguishes between systematic variations (acceptable) and localized defects like dust (unacceptable), reducing false positives from non-biological factors

Inventive Principle:
Principle #3Local quality

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach allows for the appropriate evaluation of selective binding reliability, preventing the unnecessary elimination of usable data by considering non-biological effects, thus improving the accuracy of microarray experiment results.

Implementation Method 1

mRNA or cDNA extracted and labeled from a cell to be examined may be hybridized with the spots to measure a gene expression level

Methodology Applied
Scientific EffectHybridization:

Implementation Method 2

a substance to be examined such as labeled cDNA selectively binds to a complementary DNA on the carrier

Methodology Applied
Scientific EffectSelective binding:

Data Source

PatentEP2824444B1Determination method, determination device, determination system, and program
Publication Date: 2018.04.18 TORAY INDUSTRIES INC
  • EP2824444B1 patent drawingFigure 1~2
  • EP2824444B1 patent drawingFigure 3
  • EP2824444B1 patent drawingFigure 4

AI summary

The present invention determines a position of a spot in image data obtained by imaging detection intensity in a carrier, extracts a pixel group corresponding to the spot, calculates a ratio or a difference between a median value of the detection intensity of the extracted pixel group and a median value of the detection intensity of the pixel group excluding a certain top proportion of and/or a certain bottom proportion of pixels, and determines quality of reliability based on the calculated ratio or difference and a certain reference value.