SequenceDataAnalysisforLong：时间序列数据分析.pdfVIP

Sequence Data Analysis for Long Disordered Regions Prediction in the Calcineurin Family Pedro Romero 1 Zoran Obradovic 1 A. Keith Dunker 2 promero@ zoran@ dunker@ 1 Scho ol of Electrical Engineering and Computer Science 2 Department of Bio chemistry and Biophysics, Washington State University, Pullman, WA 99164, U.S.A. Abstract Our recently reported results [14, 29, 30] provide strong support for a hypothesis that some amino acid sequences code for disordered regions rather than structured ones and that such disordered regions ar e commonly involved in function. General and family-sp ecic neural network pr edictors developed in those previous studies suggest that dierent classes of disordered regions exist. Her e, family-sp ecic data preprocessing for disorder prediction in the calcineurin (CaN) family is explored. The results show that prediction of order and disorder on CaN sequence data benets signicantly fr om the use of family-specic preprocessing, with feature extraction through principal components analysis (PCA) outperforming fe ature selection techniques, although al l methods do a go od job of discriminating CaN-specic disordered regions fr om CaN-specic ordered regions. On the other hand, for the discrimination of CaN-specic disordered regions fr om gener al (unrelated to CaN) ordered regions, fe ature selection approaches proved to be more appropriate than PCA. The results further support a hyp othesis that dierent kinds of disordered regions exist, as al l family-sp ecic disorder predictors developed in t