Journal Article

Estimation of errors introduced by confocal imaging into the data on segmentation gene expression in <i>Drosophila</i>

Ekaterina Myasnikova, Svetlana Surkova, Lena Panok, Maria Samsonova and John Reinitz

in Bioinformatics

Volume 25, issue 3, pages 346-352
Published in print February 2009 | ISSN: 1367-4803
Published online December 2008 | e-ISSN: 1460-2059 | DOI:
Estimation of errors introduced by confocal imaging into the data on segmentation gene expression in Drosophila

Show Summary Details


Motivation: Currently the confocal scanning microscopy of fluorescently tagged molecules is extensively employed to acquire quantitative data on gene expression at cellular resolution. Following this approach, we generated a large dataset on the expression of segmentation genes in the Drosophila blastoderm, that is widely used in systems biology studies. As data accuracy is of critical importance for the success of studies in this field, we took a shot to evaluate possible errors introduced in the data by acquisition and processing methods. This article deals with errors introduced by confocal microscope.

Results: In confocal imaging, the inevitable photon noise is commonly reduced by the averaging of multiple frames. The averaging may introduce errors into the data, if single frames are clipped by microscope hardware. A method based on censoring technique is used to estimate and correct this type of errors. Additional source of errors is the quantification of blurred images. To estimate and correct these errors, the Richardson–Lucy deconvolution method was modified to provide the higher accuracy of data read off from blurred images of the Drosophila blastoderm. We have found that the sizes of errors introduced by confocal imaging make up ∼5–7% of the mean intensity values and do not disguise the dynamic behavior and characteristic features of gene expression patterns. We also defined a range of microscope parameters for the acquisition of sufficiently accurate data.



Supplementary information: Supplementary data are available at Bioinformatics online.

Journal Article.  6514 words.  Illustrated.

Subjects: Bioinformatics and Computational Biology

Full text: subscription required

How to subscribe Recommend to my Librarian

Users without a subscription are not able to see the full content. Please, subscribe or login to access all content.