The Papillomavirus Episteme: a central resource for papillomavirus sequence data and analysis

Abstract

The goal of the Papillomavirus Episteme (PaVE) is to provide an integrated resource for the analysis of papillomavirus (PV) genome sequences and related information. The PaVE is a freely accessible, web-based tool (http://pave.niaid.nih.gov) created around a relational database, which enables storage, analysis and exchange of sequence information. From a design perspective, the PaVE adopts an Open Source software approach and stresses the integration and reuse of existing tools. Reference PV genome sequences have been extracted from publicly available databases and reannotated using a custom-created tool. To date, the PaVE contains 241 annotated PV genomes, 2245 genes and regions, 2004 protein sequences and 47 protein structures, which users can explore, analyze or download. The PaVE provides scientists with the data and tools needed to accelerate scientific progress for the study and treatment of diseases caused by PVs.

Figure 1.

Papillomavirus Genomic Organization. The double-stranded, circular genomes of PVs are ∼7–8 kb in length. All PVs encode the E1 and E2 replication proteins and the L1 and L2 capsid proteins. In addition to these core proteins, most viruses encode auxiliary proteins E5, E6 and E7, which manipulate host cell proliferation and cell cycle checkpoints. The URR is located between the L1 and E6 ORFs. This region contains viral promoters, enhancers and the replication origin.

Figure 2.

Genome annotation pipeline employed in the PaVE: the flowchart outlines the different steps used to curate the annotation of viral ORFs. See the main text for a detailed description.

Figure 3.

The PaVE locus viewer. The PaVE locus view shows a linear representation of the HPV16REF genome. Different annotations are displayed and can be selected. The annotated features include ORFs, spliced proteins, protein-binding site motifs and the viral URR. Upon selection, additional information is displayed in the ‘selected feature details’ window. The ‘coding region sequences’ window shows the nucleotide sequence for the selected ORF. Protein sequences can be directly compared with others in the PaVE database using the provided link to BLAST.

Figure 4.

The PaVE structure viewer. The HPV16 L1 structure (PDB No. 1DZL) is shown within the PaVE structure viewer. The PaVE structure viewer consists of five different modules. Module (A) shows basic information about the structure under investigation and includes links to the locus viewer and the original PDB file. Module (B) is constructed around a fully functional Jmol Console (www.Jmol.org), allowing manipulation of the viral structure. For some viral proteins, the structures have been solved for several homologous proteins isolated from different viral types. These alternative structures can be selected using a pull down menu in (C). A unique feature of the PaVE structure viewer is the pairwise alignment shown in (D). Under default conditions, this alignment shows the sequence of the PaVE RefClone protein to the sequence present in the PDB file. However, users can use module (E) to compare homologous sequences with the displayed structure. Statistics derived from the BLASTp alignment are provided and differences or identities can be highlighted on the alignment. Specific residues or groups of residues can be selected and highlighted on the structure.