title   
  

Prediction of neddylation sites from protein sequences and sequence-derived properties

Yavuz, Ahmet Sinan and Sözer, Namık Berk and Sezerman, Osman Uğur (2015) Prediction of neddylation sites from protein sequences and sequence-derived properties. BMC Bioinformatics, 16 (Supplement: 18). ISSN 1471-2105

[img]
Preview
PDF (Listed in DOAJ as an open access journal) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
944Kb

Official URL: http://dx.doi.org/10.1186/1471-2105-16-S18-S9

Abstract

Background Neddylation is a reversible post-translational modification that plays a vital role in maintaining cellular machinery. It is shown to affect localization, binding partners and structure of target proteins. Disruption of protein neddylation was observed in various diseases such as Alzheimer's and cancer. Therefore, understanding the neddylation mechanism and determining neddylation targets possibly bears a huge importance in further understanding the cellular processes. This study is the first attempt to predict neddylated sites from protein sequences by using several sequence and sequence-based structural features. Results We have developed a neddylation site prediction method using a support vector machine based on various sequence properties, position-specific scoring matrices, and disorder. Using 21 amino acid long lysine-centred windows, our model was able to predict neddylation sites successfully, with an average 5-fold stratified cross validation performance of 0.91, 0.91, 0.75, 0.44, 0.95 for accuracy, specificity, sensitivity, Matthew's correlation coefficient and area under curve, respectively. Independent test set results validated the robustness of reported new method. Additionally, we observed that neddylation sites are commonly flexible and there is a significant positively charged amino acid presence in neddylation sites. Conclusions In this study, a neddylation site prediction method was developed for the first time in literature. Common characteristics of neddylation sites and their discriminative properties were explored for further in silico studies on neddylation. Lastly, up-to-date neddylation dataset was provided for researchers working on post-translational modifications in the accompanying supplementary material of this article.

Item Type:Article
Additional Information:Wos Document Type: Article; Proceedings Paper / Conference: Joint 26th Genome Informatics Workshop / Asia-Pacific-Bioinformatics-Network (APBioNet) 14th International Conference on Bioinformatics (GIW/InCoB) Location: Tokyo, JAPAN Date: SEP 09-11, 2015
Uncontrolled Keywords:Neddylation, NEDD8, machine learning, support vector machines, post-translational modifications
Subjects:Q Science > Q Science (General)
ID Code:28804
Deposited By:Ahmet Sinan Yavuz
Deposited On:15 Mar 2017 15:09
Last Modified:15 Mar 2017 15:09

Repository Staff Only: item control page