Abstract and figures of paper
Automatic analysis of DNA microarray images using mathematical morphology


Jesús Angulo and Jean Serra

published in
BIOINFORMATICS, Vol. 19, no. 5, pp. 553-562 , 2003


(In order to see the figures in optimum quality the JPG files must be saved or the web page printed)
Abstract

Motivation: DNA microarrays are an experimental technology which consists in arrays of thousands of discrete DNA sequences that are printed on glass microscope slides. Image analysis is an important aspect of microarray experiments. The aim of this step is to reduce an image of spots into a table with a measure of the intensity for each spot. Efficient, accurate and automatic analysis of DNA spot images is essential in order to use this technology in laboratory routines.
Results: We present an automatic non-supervised set of algorithms for a fast and accurate spot data extraction from DNA microarrays using morphological operators which are robust to both intensity variation and artefacts. The approach can be summarised as follows. Initially, a gridding algorithm yields the automatic segmentation of the microarray image into spot quadrants which are later individually analysed. Then the analysis of the spot quadrant images is achieved in five steps. First, a pre-quantification, the spot size distribution law is calculated. Second, the background noise extraction is performed using a morphological filtering by area. Third, an orthogonal grid provides the first approach to the spot locus. Fourth, the spot segmentation or spot boundaries definition is carried out using the watershed transformation. And fifth, the outline of detected spots allows the signal quantification or spot intensities extraction; in this respect, a noise model has been investigated. The performance of the algorithm has been compared with two packages: ScanAlyze and Genepix, showing its robustness and precision.
Availability: A prototype system integrated in PDI32 (an image processing software for Windows) may be obtained from the authors on request.
  1. Introduction
  2. Systems and Methods - DNA Microarray Images
  3. Systems and Methods - Orthogonality and Image Projections
  4. Systems and Methods - Mathematical Morphology
  5. Systems and Methods - Visualisation, segmentation and quantification
  6. Algorithm - Array Orthogonal Grid

    Figure 1: Procedure for array orthogonal grid definition: (a) Decimation and low-pass filtering by 4, the initial image has a size of 1842 x 4512 pixels and the present decimated image, 460 x 1128 pixels . (b) Spot groups morphologically enhanced by means of supremum of horizontal and vertical closings. (c) Array grid obtained.

    (a)

    (b)

    (c)

    Microarray 3


    Figure 2: Examples of array gridding.

    (a)

    (b)

    Microarray 2

    Microarray 1


  7. Algorithm - Spot-Size Distribution Law

    Figure 3: Meaning of Spot-Size Distribution: (a) Examples of microarray images, array1a and array1b from the same array and array2 from other one. (b) Area extinction spectra of the examples in logarithmic scale.

    array1a

    array1b

    array2

    (a)

    (b)


  8. Algorithm - Morphological Filtering by Area Extinction Value
  9. Algorithm - Spot Orthogonal Grid
  10. Algorithm - Morphological Segmentation by Watershed Transformation
  11. Algorithm - Spot Quantification and Noise Extraction
  12. Implementation
  13. Discussion and Conclusions - Results of a comparative study

    Figure 4: Comparison of segmentation algorithms using the spot block No. 1 from Microarray 1: (a) Initial spot block. (b) Spot segmentation using the present approach. (c) Spot segmentation using ScanAlize. (d) Spot segmentation using GenePix.

    (a)

    (b)

    (c)

    (d)


    Figure 5: Comparison of segmentation algorithms using the spot block No. 1 (double) from Microarray 2: (a) Initial spot block. (b) Spot segmentation using the present approach. (c) Spot segmentation using ScanAlize. (d) Spot segmentation using GenePix.

    (a)

    (b)

    (c)

    (d)


    Figure 6: Comparison of segmentation algorithms using the spot block No. 32 from Microarray 3: (a) Initial spot block. (b) Spot segmentation using the present approach. (c) Spot segmentation using ScanAlize. (d) Spot segmentation using GenePix.

    (a)

    (b)

    (c)

    (d)


    Figure 7: Comparison of the spot quantification for the three examples (the parameter represented is the normalised mean intensity after background correction in a dye): (a) 6 random spots from Microarray 1. (b) 6 random spots from Microarray 2. (c) 5 random spots from Microarray 3.

    (a)

    (b)

    (c)


  14. Discussion and Conclusions - Conclusions and perspectives
Back to the research page

Last modification: January 1st, 2003