Plugin: Register Image - Natural Neighbor Transform for Aligning GCxGC and GCxGC-MS Data

Discuss with developer team, share useful tips, report bugs, request new features and anything else
Post Reply
cheble
Posts: 2
Joined: June 12th, 2013, 10:43 am

Plugin: Register Image - Natural Neighbor Transform for Aligning GCxGC and GCxGC-MS Data

Post by cheble » March 6th, 2018, 6:42 pm

The Natural Neighbor Transform (NIES-EPFL) plugin implements the alignment methods based on:
  • Algorithm A: Gros, J., Nabi, D., Dimitriou-Christidis, P., Rutler, R., Arey, J. S., “Robust algorithm for aligning two-dimensional chromatograms”, Anal. Chem. 2012, vol 84, p 9033-9040.
  • Algorithm B: Zushi, Y., Gros, J., Tao, Q., Reichenbach, S. E., Hashimoto, S., Arey, J. S., “Pixel-by-pixel correction of retention time shifts in chromatograms from comprehensive two-dimensional gas chromatography coupled to high resolution time-of-flight mass spectrometry”, Journal of Chromatography A 2017, vol 28, p 121-129.
The plugin allows GC Image users to use these two alignment methods directly through the Register Image action. Also, the original Matlab implementations are available at https://github.com/GCxGC/GCxGC-MS-alignment.

Installation

The plugin requires Version 2.6 or later.
  1. Download here: the NIES-EPFL Transform plugin
  2. Unzip it using File Explorer or another application.
  3. Start GC Image.
  4. From the menu, select Tools > Manage Plugins...
  5. Click on Import and then select the NIES-Alignment.plugin folder.
  6. Press OK to finish the installation. NIES-EPFL Transform will now be listed in the table.
Usage

To align chromatograms using Natural Neighbor Transform (NIES-EPFL), select Processing > Register Image... in Image or Run > Register Images in Project. The NIES-EPFL Transform option will become available in the Transform Type combo-box. More details related to the Register Image action can be found in the User Guide.

The plugin can be configured to be as same as the two original Matlab implementations as follows:
  1. Algorithm A for non-MS data
    • Choose to use Bicubic as the Interpolation Type.
  2. Algorithm B for MS and HRMS data
    • For both MS and HRMS data, choose to use Bilinear as the Interpolation Type.
    • For MS data, from Configure > Settings > Multi-Channel select By Integer Intervals for the MS Data Binning.
    • For HRMS data, from Configure > Settings > Multi-Channel select By Uniform Bin for the MS Data Binning.
There are three additional configuration settings provided for Algorithm A and B. These settings can be modified through the menu Tools > Manage Plugins... Select the plugin from the table and click on the Configure button on the tool bar.
  • Typical peak widths (Col I and Col II) in pixel units, which defaults to 1 pixel.
  • Deformation correction enabled flag in true or false, which defaults to true.
  • Linear Interpolation for 1D enabled flag in true or false, which defaults to false.
Note: In addition to Algorithm A and B, the software supports HRMS data for all interpolation types by using the GC Image's MS Data Binning settings.

Post Reply