We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Position-Specific Analysis and Prediction of Protein Pupylation Sites Based on Multiple Features.
- Authors
Xiaowei Zhao; Jiangyan Dai; Qiao Ning; Zhiqiang Ma; Minghao Yin; Pingping Sun
- Abstract
Pupylation is one of the most important posttranslational modifications of proteins; accurate identification of pupylation sites will facilitate the understanding of the molecular mechanism of pupylation. Besides the conventional experimental approaches, computational prediction of pupylation sites is much desirable for their convenience and fast speed. In this study, we developed a novel predictor to predict the pupylation sites. First, the maximum relevance minimum redundancy (mRMR) and incremental feature selection methods were made on five kinds of features to select the optimal feature set. Then the prediction model was built based on the optimal feature set with the assistant of the support vector machine algorithm. As a result, the overall jackknife success rate by the new predictor on a newly constructed benchmark dataset was 0.764, and the Mathews correlation coefficient was 0.522, indicating a good prediction. Feature analysis showed that all features types contributed to the prediction of protein pupylation sites. Further site-specific features analysis revealed that the features of sites surrounding the central lysine contributed more to the determination of pupylation sites than the other sites.
- Publication
BioMed Research International, 2013, Vol 2013, p1
- ISSN
2314-6133
- Publication type
Article
- DOI
10.1155/2013/109549