©1996-2009 All Rights Reserved. Online Journal
of Bioinformatics. You may not store these pages in any form except for
your own personal use. All other usage or distribution is illegal under international
copyright treaties. Permission to use any of these pages in
any other way besides the before mentioned must be gained in writing from the
publisher. This article is exclusively copyrighted in its entirety to
OJB publications. This article may be copied once but may not be reproduced or
re-transmitted without the express permission of the editors.
OJB©
Online Journal of Bioinformatics©
Volume
6 : 51-64, 2005
Höglund A, Dönnes P, Adolph HW, Kohlbacher O
Höglund A, Dönnes P, Adolph HW, Kohlbacher O. From prediction of subcellular localization to
functional classification: Discrimination of DNA-packing and other nuclear
proteins, Online J Bioinformatics, 6 : 51-64, 2005 Subcellular localization is related to protein function.
Computational methods have shown that different chemical environments in the
cell lead to evolutionary adaptation of amino acid composition for cytoplasmic, extracellular,
mitochondrial, and nuclear proteins. In this study, the division of nuclear
proteins into functional groups and, the influence of sequence homology in the
assessment of prediction accuracy was
investigated. Results showed that histones and histone-like proteins, all involved in DNA-packing in
eukaryotic cells could be separated from other proteins with high statistical
significance. The proteins are a distinct separate group among the nuclear
proteins, extending the classification of subcellular
localization with functional annotation. On homology-reduced data the clear
separation of proteins from different localizations as reported in previous
studies was not found. A nearest neighbour classifier performs even better than
a machine learning approach on the original data set. The findings suggest that
performance should be evaluated at different levels of sequence homology in
order to provide a measure of the robustness of the method.
KEYWORDS: protein, subcellular
localization, nuclear protein, machine learning