Issue No. 01 - January/February (2012 vol. 9)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/TCBB.2011.85
Yun Xu , Anhui Province Key Lab. of High Performance Comput., Univ. of Sci. & Technol. of China, Hefei, China
Da Teng , Anhui Province Key Lab. of High Performance Comput., Univ. of Sci. & Technol. of China, Hefei, China
Yiming Lei , Anhui Province Key Lab. of High Performance Comput., Univ. of Sci. & Technol. of China, Hefei, China
The rapid growth of scientific literature calls for automatic and efficient ways to facilitate extracting experimental data on protein phosphorylation. Such information is of great value for biologists in studying cellular processes and diseases such as cancer and diabetes. Existing approaches like RLIMS-P are mainly rule based. The performance lays much reliance on the completeness of rules. We propose an SVM-based system known as MinePhos which outperforms RLIMS-P in both precision and recall of information extraction when tested on a set of articles randomly chosen from PubMed.
Proteins, Substrates, Data mining, Dictionaries, Abstracts, Databases, Bioinformatics
Yun Xu, Da Teng and Yiming Lei, "MinePhos: A Literature Mining System for Protein Phoshphorylation Information Extraction," in IEEE/ACM Transactions on Computational Biology and Bioinformatics, vol. 9, no. 1, pp. 311-315, 2011.