Tutorial of Jenner Predict Server:
The "Jenner-Predict" server has been developed to provide credible vaccine candidates and information related to their immunogenic potential, conservation and autoimmunity in host. The server is based on the principle that non-cytosolic proteins having functions important in host-pathogen interactions and disease establishment could be potential vaccine candidates. Since the server prioritizes few numbers (3-5% proteins of a proteome) of protein vaccine candidates (PVCs), user may take few of these proteins and evaluate their immune response for subunit vaccine development.
Section A: Job submission in Jenner-Predict server
To submit a job, please open home page of Jenner-Predict server available at URL: http://188.8.131.52/vaccine/home.html. The Server, Jenner-Predict, provides three simple ways for submission of jobs (Figure 1). If user wants to paste protein sequences then sequence(s) can be directly pasted to the text box. If user wants to upload file containing protein sequences then he/she should click on from "Browse/choose file" button. In both the above cases, 'paste' sequence(s) and upload file, user needs to provide gram information by clicking on gram negative or positive. If user wants to give whole proteome of a bacterium as input then he/she should select the name of the organisms from list.
Figure T1: Pictorial representation of job submission in Jenner-Predict server
When a job is submitted successfully (just after clicking submit button), the server will generate a link html page where output or status will be available. Since jobs are processed in a queue, user should bookmark this link to track status or output of his/her job.
The Job 'status' may be one of the following:
After completion of job, results will be available on provided link & it will not removed before three weeks from the server.
Section B: Result output Jenner-Predict Server
Result of Jenner-Predict server includes output of all analysis done on a given protein sequences. Result is provided in tabular form. The format of result for a given job is in following figure:
Figure 3: Sample output of Jenner-Predict server
The information provided in different columns is given as follows:
1. Serial number
2. Gene Id
3. Cellular localization of protein
4. No. of transmembrane helix/helices
5. Pfam domain ID
6. No. of IEDB T-cell epitope match(s)
7. No. of IEDB B-cell epitope match(s) ((hyperlinks on 6 & 7) showing alignment(s) & matching positions are provided)
8. Autoimmunity information through 35% identical match in 80 AA lengths
9. Autoimmunity information through the number of 9-mer identical match in an alignment ((hyperlinks on 8 & 9) showing alignment(s))
10. Conservation in pathogenic/nonpathogenic strains (x/y/z) where x: all strains of that bacteria, y: pathogenic strains of that bacteria, and non-pathogenic strains of that bacteria
Section 3: Output Analysis and Prioritization of PVCs
Proteome Data and Software:
Last five columns (analysis part) provide information for analyzing vaccine potential of predicted PVCs. Details of the analysis columns are given as follows.
T-cell epitope (TCE) Blast match against the PVC has two numeric values which are separated by "/", for example "x/y" where "x" is the number of total IEDB epitope match(s) and 'y' is the number of non-overlapping IEDB epitope match(s) on different positions in PVC. Hyperlink associated with this column provides alignments of T-cell epitopes with the PVC. The hyperlink also contains IEDB epitope ID, PubMed ID, protein accession ID, etc. related to that T-cell epitope (Figure TA1). The PubMed ID may be used to find experimental information of the TCE. The epitope ID may be used for IEDB database search to get details data regarding that epitope.
Figure TA1: Representation of alignment of T-Cell epitope with predicted PVC.
B-cell epitope (BCE) Blast match(s) against the PVC has two numeric values separated by "/", for example "x/y" where "x" is the number of total IEDB epitope match(s) and 'y' is the number of non-overlapping IEDB epitope match(s) on different positions in PVC. Hyperlink associated with this column provides alignments of B-cell epitope(s) with the PVC. As in the case of TCE (column 6), hyperlink also contains IEDB epitope ID, PubMed ID, protein accession ID, etc. related to the BCE. In columns 6 & 7, if any epitope is having exact match on PVC then that entire row displayed with white background. A sample BCE match against PVC has a similar representation as in Figure TA1.
This column provides number of human protein(s) similar to the PVC with criteria of 35% identity in 80 amino acid length. Hyperlink associated with this column provides alignments of predicted PVC with human protein(s)(Figure TA2).
Figure TA2: Human homolg of PVC with exact 9 or more amino acid continuous match
This column provides number of exact 9 or more amino acid continuous match in an alignment of predicted PVC with sequences of human proteome. Hyperlink on it provides, exact matching length, and human sequence information (protein id, name, etc.) (Figure TA3).
Figure TA3: Human homolg of PVC with exact 9 or more amino acid continuous match
Conservation in pathogenic/nonpathogenic strains (x/y/z) where x: all strains of that bacteria, y: pathogenic strains of that bacteria, and z: non-pathogenic strains of that bacteria.. Hyperlink associated with this blocks porovide tabular representation of strains names in different column (Figure: TA4)
Figure TA4: Condervaton of Pvc within Pathogenic and Non-pathogenic strains
Although Jenner-Predict server provides lesser number of proteins as vaccine candidates (PVCs), still output of the server is prioritized on the basis of analysis (columns 6-10) so that user may consider few proteins for evaluation of their vaccine candidate potential. The server ranks PVCs in the result table on the basis of two analyses: Predicted PVCs are de-prioritized according to number of human homolgs match of PVC (only the more stringent exact 9 or more amino acid continuous match in an alignment (column 9) is considered) and higher prioritization of PVC is proportional to number of IEDB epitope match(s). The former is based on finding that similarity with human protein which may lead to autoimmunity/weak immune response and therefore, more number of similar proteins with human is ranked lower in the result table. While the latter indicate that more number of IEDB epitope matches then higher will be the probability of inducing immune response.