Research On Sequence-structure Anlalysis And Mutation Study In HPV Types Of Cervical Cancer

Cervical cancer is the second most common malignancy in women worldwide,which is second to breast cancer.A large number of basic and clinical studies have found that high-risk HPV persistent infection is one of the key factors for cervical cancer.In recent years,China has reported some mutations in HPV high-risk types,there are many mutation patterns in the cervix that infect high-risk HPV type.Now,many HPV studies focused on the HPV sequences and ignored the clinical mutation information.In this dissertation,we studied the protein structure comparison and prediction model of HPV types,and discussed the relationship between the mutations and protein sequence/structure motifs.The main work is represented as follows:1.We summarized the basic knowledge of human papillomavirus(HPV)and cervical cancer,including human papillomavirus type,structure,function,human papillomavirus-related diseases,the relationship between human papillomavirus and cervical cancer,which provides powerful theoretical foundation for the following research.2.We schemed out a new method to analyze the similarity/dissimilarity of the protein structures based on Markov random fields.We calculated the revised contact map matrix using the distribution of the distance matrix instead of an established threshold and constructed Markov random fields with different cliques.We then calculated the conditional probabilities of the Markov random fields to compare the different protein structures.The results indicate that the proposed method exhibits a strong ability to detect the similarities/dissimilarities among the conformation of different cyclic peptides and protein structures.We also found that the arpha-C,oxygen O and N allow us to extract more conserved structures of the proteins,and Markov random fields with 2-point cliques are more efficient.3.We proposed a prediction of high-risk types of human papillomaviruses based on amino acid characteristics.We introduced the physicochemical properties of amino acids to reduce amino acids and extracted six kinds of characteristic information of protein.With help of support vector machine,we constructed prediction model of high-risk types of human papillomaviruses.Experiment results indicate that the proposed model is more effective to identify high-risk HPV and low-risk HPV.In addition,we found that it is better to select beta physicochemical properties to reduce amino acids when using E5,E6,E7,L1 and L2 proteins to classify HPV types.If E1,E2,E4,E5 and E7 proteins were used to classify HPV types,PRse AAC will be better.It is better to select RTCD when using E6,L1 and L2 proteins to classify HPV types.4.We discussed the relationship between the mutations and protein sequence/structure motifs.We first collected a large number of mutations of domestic cervical cancer HPV by literature searching,and studied the relationship between the mutations and the sequence conserved regions or structure conserved domains.The results indicate that there are 134,86 and 166 mutations in E6,E7 and L1 proteins,which are more than other proteins.As for low-risk type E6,there is 9 of 11 mutations in the p53 protein binding region or antigenic region.But as high risk type E6,there is 49 of 91 mutations in the p53 protein-binding region or in the antigenic determinant region.We also found that the percentage of mutations in the E7 domain is high,with more than 93% of the mutations in functional domain.