Critical evaluation of quantitative colocalization analysis in confocal fluorescence microscopy

Spatial colocalization of fluorescently labeled proteins can reveal valuable information about proteinprotein interactions. Compared to qualitative visual interpretation of dual color images, quantitative colocalization analysis (QCA) provides more objective evaluations to the degree of colocalizati...

Full description

Saved in:
Bibliographic Details
Published in:Interdisciplinary sciences : computational life sciences 2012-03, Vol.4 (1), p.27-37
Main Authors: Wu, Yong, Zinchuk, Vadim, Grossenbacher-Zinchuk, Olga, Stefani, Enrico
Format: Article
Language:English
Subjects:
Biological
Biomedical and Life Sciences
Computational Biology/Bioinformatics
Computational Science and Engineering
Computer Appl. in Life Sciences
Computer Simulation
Confocal
Density
Fluorescence
Fluorescence microscopy
Health Sciences
Life Sciences
Mathematical and Computational Physics
Medicine
Microscopy, Confocal - methods
Microscopy, Fluorescence - methods
Molecular structure
Position (location)
Proteins
Proteins - metabolism
Reproducibility of Results
Statistics for Life Sciences
Subcellular Fractions
Theoretical
Theoretical and Computational Chemistry
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Spatial colocalization of fluorescently labeled proteins can reveal valuable information about proteinprotein interactions. Compared to qualitative visual interpretation of dual color images, quantitative colocalization analysis (QCA) provides more objective evaluations to the degree of colocalization. However, the finite resolution power of microscopes and the spatial patterns of intracellular structures may compromise the reliability of many classical QCA methods. In this paper, we discuss the strength and weakness of some mostly used QCA methods. By studying their applications on computer-simulated images and biological images, we show that classical pixel intensity based QCA methods are often vulnerable to coincidental overlapping among resolution elements (resel) distributions and thus not suitable to images with high molecular density or with low resolution. Also, many QCA methods can mistakenly regard long range correlation as colocalization due to protein localization in intracellular structures. The newly developed protein-protein index (PPI) approach is able to reduce the influence from resel overlapping and spatial intracellular pattern compared to previous methods, significantly improving the reliability of QCA.
ISSN:1913-2751
1867-1462
DOI:10.1007/s12539-012-0117-x