Prediction of essential proteins based on gene expression programmingReport as inadecuate

Prediction of essential proteins based on gene expression programming - Download this document for free, or read online. Document in PDF available to download.

BMC Genomics

, 14:S7

First Online: 01 October 2013DOI: 10.1186-1471-2164-14-S4-S7

Cite this article as: Zhong, J., Wang, J., Peng, W. et al. BMC Genomics 2013 14Suppl 4: S7. doi:10.1186-1471-2164-14-S4-S7


BackgroundEssential proteins are indispensable for cell survive. Identifying essential proteins is very important for improving our understanding the way of a cell working. There are various types of features related to the essentiality of proteins. Many methods have been proposed to combine some of them to predict essential proteins. However, it is still a big challenge for designing an effective method to predict them by integrating different features, and explaining how these selected features decide the essentiality of protein. Gene expression programming GEP is a learning algorithm and what it learns specifically is about relationships between variables in sets of data and then builds models to explain these relationships.

ResultsIn this work, we propose a GEP-based method to predict essential protein by combing some biological features and topological features. We carry out experiments on S. cerevisiae data. The experimental results show that the our method achieves better prediction performance than those methods using individual features. Moreover, our method outperforms some machine learning methods and performs as well as a method which is obtained by combining the outputs of eight machine learning methods.

ConclusionsThe accuracy of predicting essential proteins can been improved by using GEP method to combine some topological features and biological features.

Download fulltext PDF

Author: Jiancheng Zhong - Jianxin Wang - Wei Peng - Zhen Zhang - Yi Pan


Related documents