19th IEEE Symposium on Computer-Based Medical Systems (CBMS'06) (2006)
Salt Lake City, Utah
June 22, 2006 to June 23, 2006
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/CBMS.2006.103
L. Palopoli , Universita della Calabria, Italy
S. E. Rombo , Universita "Mediterranea" di Reggio Calabria, Italy
G. Terracina , Universita della Calabria, Italy
G. Tradigo , ICAR-CNR, Italy
P. Veltri , Universita "Magna Gr?cia" di Catanzaro, Italy
Identifying protein secondary structures is a difficult task. Recently, a lot of software tools for protein secondary structure prediction have been produced and made available on-line, mostly with good performances. However, prediction tools work correctly for families of proteins, such that users have to know which predictor to use for a given unknown protein. We propose a framework to improve secondary structure prediction by integrating results obtained from a set of available predictors. Our contribution consists in the definition of a two phase approach: (i) select a set of predictors which have good performances with the unknown protein family, and (ii) integrate the prediction results of the selected prediction tools. Experimental results are also reported.
P. Veltri, G. Tradigo, S. E. Rombo, L. Palopoli and G. Terracina, "JSSPrediction: a Framework to Predict Protein Secondary Structures Using Integration," 19th IEEE Symposium on Computer-Based Medical Systems (CBMS'06)(CBMS), Salt Lake City, Utah, 2006, pp. 931-935.