Amongst the existing 200 strains of HPV, 150 have been sequenced as of now, and their data available at the National Center for Biotechnology Information (NCBI). Genome and proteome information of those viral strains was retrieved from NCBI. Besides PubMed, various other online resources and published literature were also explored for validating genomic, proteomic, as well as strain and disease-associated information on HPV strains. HPV strain-specific information, such as strain name, sequencing status, sequencing centre, NCBI accession ID, associated disease information with references, genome statistics (GC%, AT%, A, T, G, C count, genes, and proteins), etc., were curated from various online resources, and protein parameters (length, molecular weight, isoelectric point) were calculated using ExPASy ProtParam [4].

MODELLER9v10 [5] and the SWISS-MODEL [6] server were used for protein structure prediction. The stereochemistry of each protein was evaluated through PROCHECK [7] analysis, available at the RCSB validation server (http://deposit.rcsb.org/validate/), and validated using ProSA-web [8] (http://prosa.services.came.sbg.ac.at/prosa.php).

Human Papillomavirus Proteome Database (hpvPDB), the relational reserve, was developed using Microsoft SQL Server 2005 as the back end. The website is powered by XAMPP (Windows Version 1.7.3). HTML, JavaScript, and CGI-PERL-based web interfaces were employed to execute SQL queries. The curated data and related information were stored in tables. The application layer, the web interface, and the backend relational tables were integrated using CGI-PERL. The overall architecture of hpvPDB is shown in Fig. 1.

hpvPDB interfaces are made to help the users for easy navigation and information retrieval. Home, About, Tools, Search, and Advanced Search interfaces can be explored to obtain strain- and protein-specific information. User can access the meta information about different strains using a search box. Reserve comprises the strain-specific detailed informationon its name, sequencing status, submission details, date of submission, NCBI IDs, disease types and subtypes, type of DNA, genome length, molecular weight, nucleotide composition (A, T, G, C, AT, GC content), number of genes and proteins, and protein list. A genome map of each strain obtained by Geneious 5.4.4 software (available from http://www.geneious.com/) is also integrated in this page. Users, through an advanced search option, can precisely access the Genome and Proteome information separately by selecting HPV genome or HPV proteome. Each protein entry comprises protein overview (name, locus, function, etc.), protein sequence information (amino acid sequences with NCBI accession number with provision for direct protein BLAST [9] against NCBI nr database), protein parameters (length, molecular weight, theoretical isoelectric point [pI], amino acid composition, etc.), protein structure (predicted 3D structure by homology modeling viewed by Jmol (available from http://www.jmol.org/) [10] with the JAVA platform, Ramachandran plot obtained by PROCHECK and Z-score and Energy plot obtained by ProSA-Web. hpvPDB platform also provides a phylogeny analysis tool to perform multiple sequence alignment and phylogenetic tree construction of selected HPV proteins using the Phylogeny. fr web service [11].

The original Human Papillomaviruses Database was developed and hosted by the Los Alamos National Laboratory (LANL) between 1994 and 1999 with funding from the National Institute of Allergy and Infectious Diseases (NIAID) [12]. 'Human Papillomaviruses: A Compilation and Analysis of Nucleic Acid and Amino Acid Sequences' contains four annual data books of papillomavirus information published in both paper and electronic form (1994, 1995, 1996, and 1997) but has not been updated since 1997 [12]. This contains nucleotide sequences of few HPV strains and other papillomaviruses, amino acid and nucleotide sequence alignments, analysis, related host sequences, and database communication. We did not find any structural information in that database. In hpvPDB, along with updated protein sequence information, genome and protein structure information is also provided.