Title :
Identification of mammalian GPI-anchored proteins based on amino acid propensities in their GPI attachment signal sequences
Author :
Mukai, Yuri ; Sasaki, Takanori ; Oura, Osamu ; Ikeda, Masami
Author_Institution :
Dept. of Electron. & Bioinf., Meiji Univ., Kawasaki, Japan
Abstract :
Attachment of glycosylphosphatidylinositol (GPI) is one of the most important posttranslational modifications, playing an important role in vital eukaryote activities. GPI-anchored proteins (GPI-APs) are characterized by a pro-peptide of hydrophobic residues and small amino acid residues near the GPI-anchoring site. Here, we describe a new method for identifying GPI-APs based on hydropathy profiles and a position-specific scoring matrix (PSSM). First, the sequences of mammalian GPI-APs from the UniProt Knowledgebase/Swiss-Prot protein sequence database release 54.0 were scanned for their average hydropathy with several different window sizes. Hydrophobic regions were observed not only in the signal-peptide but also in the pro-peptide at the C-terminus. Non-GPI-anchored proteins (non-GPI-APs) with similar hydropathy profiles to those of the GPI-APs were used as the negative control. The sequences were aligned according to the residue sizes in the C-terminal region, and the position-specific amino acid propensities were analyzed according to their alignment positions in both the GPI-APs and the negative controls. The PSSM was devised using each amino acid propensity and a matching score was estimated for each dataset. The accuracy achieved in discriminating between GPI-APs and the negative controls was evaluated, and the GPI-APs were detected with 96.5% sensitivity and 97.3% specificity on a self-consistency test and with 85.0% sensitivity and 92.7% specificity on a 4-fold cross-validation test.
Keywords :
cellular biophysics; deductive databases; macromolecules; medical signal processing; molecular biophysics; proteins; GPI attachment signal sequence; SwissProt protein sequence database; UniProt Knowledge base; amino acid propensity; glycosylphosphatidylinositol; hydropathy profile; mammalian GPI anchored protein identification; position specific scoring matrix; posttranslational modification; vital eukaryote activity; Accuracy; Amino acids; Bioinformatics; Genomics; Indexes; Proteins; Sensitivity; GPI-anchored protein (GPI-AP); omega-site; position-specific scoring matrix (PSSM); pro-peptide; signal-peptide;
Conference_Titel :
Trendz in Information Sciences & Computing (TISC), 2010
Conference_Location :
Chennai
Print_ISBN :
978-1-4244-9007-3
DOI :
10.1109/TISC.2010.5714617