2D Gel Image Analysis

CSIRO research

Two dimensional (2D) electrophoresis gel images can be used for identifying and characterising many forms of a particular protein encoded by a single gene. 2D gel image analysis is very important in the field of proteomics. In order to carry out gel image analysis, one first needs to accurately detect and measure the protein spots in a gel image. These can be achieved using our advanced image segmentation techniques, such as mathematical morphology and seeded region growing methods. The next most pressing problem is one of image registration – ensuring that identical proteins in different gels are recognised as being identical. This problem is complicated by variation from gel to gel. Any given protein may not necessarily be in the same physical position or even exactly the same shape. Sophisticated algorithms have been developed for such registration purposes.

Conventional approaches to gel analysis require the following steps:

  • Spot detection on each gel
  • Spot matching between gels
  • Spot quantification and comparison

Existing software attempts to automate all steps as much as possible, but errors in the detection and matching stages are common. This means that gel analysis requires a significant level of operator interaction, which is very time consuming.

CSIRO’s approach is to explore techniques to improve the level of automation and therefore increase the rate of gel analysis. The starting point for these techniques is the registration of gels without user interaction. We have developed a number of approaches to this problem, examples of which are shown below.

Once gels are registered it is possible to produce much more reliable approaches for the other steps.

The images immediately below show a pair of gels before and after registration and filtering. This registration is performed without user interaction. One gel is shown in red, the other in blue. The grids in the lower image show the warping function that has been used to map the second image to the first one.

weld_pair

pair_with_grid

protein_regp

 Another example of protein spot detection and registration techniques developed by our Group. The small highlighted area shows the sensitivity of the spot detection process, with each spot shown in red within the yellow selection area. The yellow crosses indicate the locations of a subset of the spots detected, and the green lines show the matching relationship of the detected spots between the two images. The two input images are obtained from the NCI Flicker Web server at: http://www.lecb.ncifcrf.gov/flicker/.

Work with Proteome Systems Limited

CSIRO Mathematics, Informatics and Statistics (CMIS) and Proteome Systems Ltd (PSL) have collaborated on several projects using the proprietary image analysis functions developed by CMIS. In both the applications described below, CMIS and PSL have developed proprietary software to facilitate the interface between PSL’s protein array technology and automated technologies for protein analysis.

Image analysis software

The features of Proteome Systems’ powerful gel image analysis software include:

  • the acquisition and automatic archiving of high resolution images of PSL’s GelChipTM products.

  • automatic detection of protein spots with optional user “click-and-pick” facilities for additional protein spot editing.

  • “trend analysis” of protein spots in a gel.

The Xcise

The “Xcise” is an integrated instrument for the processing of PSL’s GelChipTM products. The system integrates a scanner for image capture and analysis, an automated cutting tool and an 8-probe liquid handling unit.

The powerful image analysis software described previously is used to control the Xcise, allowing for automated protein spot detection and excision.

Precision XY motor control directs the cutter head and delivers the excised gel pieces to standard 96-well microtitre plates. Washing, destaining, enzyme digestion, incubation, peptide extraction, cleanup, and spotting onto standard 96- and 384- sample MALDI-TOF target plate can all be performed in situ automatically.

Commercial Partners:

Proteome Systems Ltd.

http://www.proteomesystems.com