PVS: a web server for protein sequence variability analysis tuned to facilitate conserved epitope discovery

Abstract

We have developed PVS (Protein Variability Server), a web-based tool that uses several variability metrics to compute the absolute site variability in multiple protein-sequence alignments (MSAs). The variability is then assigned to a user-selected reference sequence consisting of either the first sequence in the alignment or a consensus sequence. Subsequently, PVS performs tasks that are relevant for structure-function studies, such as plotting and visualizing the variability in a relevant 3D-structure. Neatly, PVS also implements some other tasks that are thought to facilitate the design of epitope discovery-driven vaccines against pathogens where sequence variability largely contributes to immune evasion. Thus, PVS can return the conserved fragments in the MSA-as defined by a user-provided variability threshold-and locate them in a relevant 3D-structure. Furthermore, PVS can return a variability-masked sequence, which can be directly submitted to the RANKPEP server for the prediction of conserved T-cell epitopes. PVS is freely available at: http://imed.med.ucm.es/PVS/.

Figure 1.

PVS output. The figure shows a composition with the possible outputs of PVS. Results were obtained using an MSA corresponding to the HIV1 glycoprotein gp120 (residues 31–183 in gp160 from HIV-1 strain H2XB2). The MSA was generated from 359 representative sequences of the HIV-1 clades A (73), B (85), C (85), D (51) and 01_AE (65) using the program MUSCLE (29). The MSA is available at http://imed.med.ucm.es/PVS/supplemental/gp120_aln.html. The sequence variability was computed using the ‘Shannon’, ‘Simpson’ and ‘Wu-Kabat’ methods, and from the ‘sequence variability options’, a reference ‘consensus sequence’ and the default ‘variability threshold of 1.0’ were selected. (A) ‘Variability plot’. Users can change the variability metric (‘Shannon’, ‘Simpson’ and ‘Wu-Kabat’) by clicking on the relevant links. (B) ‘Variability masked sequence’. The sequence is returned in FASTA and T-cell epitope predictions can be obtained by clicking on the ‘Run Epitope Prediction’ bottom. (C) ‘Conserved fragments with no variable residues’. In this example, a ‘minimal fragment length’ of eight was selected. (D) ‘Structural variability mapping’. Sequence variability in the alignment was mapped onto the 3D-coordinates of gp120 (chain G of PDB 1RZK). The output allows the visualization of the variability in several user-selected renderings of the 3D structure. PVS can also display a graph of the protein sequence with the conserved fragments shown in blue. By clicking on a fragment, the user will locate it on the 3D-structure as shown in (E) with fragment 2. The output used to make this figure is available at: http://imed.med.ucm.es/PVS/supplemental/gp120_pvs.html.