Using the concept of Chou's pseudo amino acid composition for risk type prediction of human papillomaviruses.

High-risk types of human papillomaviruses (HPVs) are the etiological agents in nearly all cases (99.7%) of cervical cancer, and the HPV E6 protein is one of the two viral oncoproteins which is expressed in virtually all HPV-positive cancers. Therefore, classifying the risk type of HPVs is very useful and necessary for diagnosis and remedy of cervical cancer. To predict and to classify the risk types of HPV by bioinformatics analysis, 96 E6 protein sequences from available databases were obtained. To investigate the risk type of these sequences, PseAAC server, ROC curves and statistical analysis were applied. Our classification was based on some characters of HPV E6 proteins, such as hydrophobicity, hydrophilicity, side chain mass, PK of the alpha-COOH group, PK of the alpha-NH3(+) group and PI at 25 degrees C. Risk type of 4 unknown HPV types and 25 non-reported HPV types were also predicted. These results show that bioinformatics based theoretical approaches can direct and simplify experimental studies. Copyright 2009 Elsevier Ltd. All rights reserved.