Premium
HERMeS: A second generation approach to the automatic analysis of two‐dimensional electrophoresis gels Part I: Data acquisition
Author(s) -
Vincens Pierre,
Paris Nicolas,
Pujol JeanLuc,
Gaboriaud Christine,
Rabilloud Thierry,
Pennetier JeanLouis,
Matherat Philippe,
Tarroux Philippe
Publication year - 1986
Publication title -
electrophoresis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.666
H-Index - 158
eISSN - 1522-2683
pISSN - 0173-0835
DOI - 10.1002/elps.1150070802
Subject(s) - computer science , histogram , algorithm , entropy (arrow of time) , image (mathematics) , covariance , gaussian , artificial intelligence , computer vision , mathematics , statistics , physics , quantum mechanics
In this series of papers, we describe the computer system we have developed for two‐dimensional (2‐D) gel analysis. Here the acquisition part of the system is described and a set of criteria able to measure the overall performance of the system is presented. These criteria are essentially based on frequency domain considerations using Fourier transforms. We show that the transfer function characteristics of the input system must be carefully checked to avoid any image distortion; we propose an easily computed criterion to obtain setting conditions giving a correct system response. These features are discussed considering the specificity of the 2‐D gel images that can be well modeled by a multi‐Gaussian surface. Then the visualization device and its main abilities to treat and modify the aspect of the images are presented. Statistical functions using second‐order image histograms are introduced. They can be used to look at the general aspect of the images, as a global and accurate criterion for quality checking. For this purpose, covariance, entropy and inertia have been found to be most suitable. Second‐order histograms give additional information on image anisotropy and can direct the user towards the convenient values of the parameters for the subsequent treatments. Coding is considered from both the point of view of image compression and image sending throughout a computer network. An efficient code protecting all the image informations is suggested and discussed.