Identifying antigenicity-associated sites in highly pathogenic H5N1 influenza virus hemagglutinin by using sparse learning

Zhipeng Cai, Mariette F. Ducatez, Jialiang Yang, Tong Zhang, Li Ping Long, Adrianus C. Boon, Richard J. Webby, Xiu Feng Wan

Research output: Contribution to journalArticle

31 Scopus citations

Abstract

Since the isolation of A/goose/Guangdong/1/1996 (H5N1) in farmed geese in southern China, highly pathogenic H5N1 avian influenza viruses have posed a continuous threat to both public and animal health. The non-synonymous mutation of the H5 hemagglutinin (HA) gene has resulted in antigenic drift, leading to difficulties in both clinical diagnosis and vaccine strain selection. Characterizing H5N1's antigenic profiles would help resolve these problems. In this study, a novel sparse learning method was developed to identify antigenicity-associated sites in influenza A viruses on the basis of immunologic data sets (i.e., from hemagglutination inhibition and microneutralization assays) and HA protein sequences. Twenty-one potential antigenicity-associated sites were identified. A total of 17 H5N1 mutants were used to validate the effects of 11 of these predicted sites on H5N1's antigenicity, including 7 newly identified sites not located in reported antibody binding sites. The experimental data confirmed that mutations of these tested sites lead to changes in viral antigenicity, validating our method.

Original languageEnglish
Pages (from-to)145-155
Number of pages11
JournalJournal of Molecular Biology
Volume422
Issue number1
DOIs
StatePublished - Sep 7 2012

Keywords

  • H5N1
  • antibody binding site
  • antigenic drift
  • antigenicity
  • sparse learning

Fingerprint Dive into the research topics of 'Identifying antigenicity-associated sites in highly pathogenic H5N1 influenza virus hemagglutinin by using sparse learning'. Together they form a unique fingerprint.

  • Cite this