Computational Analysis of Signal Peptide-Dependent Secreted Proteins in Saccharomyces cerevisiae

摘要

Computer based software such as the SignalP v3.0, TargetP vl.01, big-PI predictor and TMHMM v2.0 were combined to predict the signal peptides, and the signal peptide-dependent secreted proteins among the 6 700 ORFs in genome of Saccharomyces cerevisiae. The results showed that 163 proteins were the secreted ones containing signal peptides, and they were secreted via Sec pathway. Among the 163 predicted secreted proteins, the signal peptides of 47 secreted proteins included only the H-domain and C-domain, without N-domain, but the signal peptides of other 116 secreted proteins included all the three domains. There were differences in the constitution of signal peptides between the secreted proteins of S. cerevisiae and of Candida albicans, but the length and amino acids types of their signal peptides were similar in general. Few of the same signal peptides occurred in the secreted proteins of S. cerevisiae genome, and the homology could be compared among the secreted proteins with the same signal peptides. The BLAST 2 SEQUENECES and CLUSTAL W were used to align the two protein sequences and multi-protein sequences, respectively. The alignment result indicated that homology of these sequences with the same signal peptide was very highly conservative in amino acid of complete gene. The effect of the signal peptides in S. cerevisia on expression of foreign eukaryotic secreted proteins is discussed in this paper.