Andrej Ridzik, Broňa Brejová. Neuropeptide Recognition by Machine Learning Methods. In V. Kurkova, L. Bajer, V. Svatek, ed., Information Technologies - Applications and Theory (ITAT), number 1003 in CEUR-WS, pp. 72-78, Jasna, Slovakia, 2014.
Download preprint: not available
Download from publisher: http://ceur-ws.org/Vol-1214/72.pdf
Related web page: not available
Bibliography entry: BibTeX
Thanks to advances in DNA sequencing and bioinformatics methods, for many species we know their genomic sequence as well as sequences of proteins produced by their cells. However, the exact function of those proteins is often unknown. The goal of our work is to automatically recognize neuropeptides, special proteins involved in communication between neurons. Neuropeptides are created in a cell from longer protein precursors, and our goal is to determine, if a given protein is a likely precursor and if yes, which of its parts will serve as neuropeptides. Existing methods solve only partial aspects of this problem. Our more comprehensive system uses a combination of support vector machines and conditional random fields.