Documentation Vaxi-DL Website

1. Accessing Vaxi-DL

Vaxi-DL is available online at http://vac.kamalrawal.in. The sequence submission form for the current version of the program is located at http://vac.kamalrawal.in/vaxidl

2. Submitting a Sequence for Analysis on the WWW

2.1 Sequence Submission:

The sequence submission form can be found at http://vac.kamalrawal.in/vaxidl. One or more sequences can be pasted into the text box, or the "Choose file" option can be used to analyze a file of one or more sequences stored on your computer. Please note that the following restrictions are enforced:

1. Minimum 30 residues per sequence are required
2. Maximum 4000 per sequence residues are allowed (Bacteria only)
3. Maximum 200000 total residues in an input are allowed (Bacteria only)
4. Per request 100 protein sequences are allowed. Rest are ignored.
5. Only protein sequences are allowed
6. Only FASTA format or UniProt ID is allowed

2.2 Choose ensemble model approach:

We created two different models with a different ensemble approach.

In the consensus-based approach, we have 5 individual models and the consensus of all 5 individual models is considered to be our final result. Since in this method, we consider the final classification of individual models i.e., 0 or 1 (non-vaccine candidate or vaccine candidate) hence probabilities get lost and that is why we don't generate a probability percentage for this approach.

In the average-based approach, we have 5 individual models and the average of all 5 individual models output is considered to be as the final result. Since in this method, we average the raw output of each model and hence the final probabilities is also the average of each probability and that is why we generate a probability percentage for this approach.

2.3 Receiving results:

There are two ways results from Vaxi-DL can be received, via email (contains result link) or wait screen display (wait till analysis is complete and then automatically user is redirected to result link). All the submissions on our server must provide an email address for result notification.

In an email, we provide you the result link if we successfully analysed your input or an explanation if we unable to parse your input or server encountered an exception during your analysis.

2.4 Acceptable Formats:

Vaxi-DL requires that a PROTEIN sequences be submitted in FASTA format.

A sequence within a FASTA sequence file consists of three parts:

• A title line, which must begin with a '>' symbol, and may be followed by any type of text

• A newline character at the end of the title line

• The sequence itself, which continues until the end of file or the next '>' is reached

For an example of FASTA format - see this.