LBtope Help Page

Antigen sequence Multiple peptides Peptide mutatants
Antigen sequence

LBtope server discriminate the linear B cell epitopes and non-epitopes for a given protein sequence. The LBTOPE server uses the SVM based models using dipeptide composition generated from the query sequence(s). The overall accuracy of this server is ~ 81%.

Sequence Input and Format

Amino acid sequences must be entered in the standard FASTA format. Sequence(s) can be pasted or can be given as a file. All the non standard characters like [*&^%$@#!()_+~=;'",<>?.\|} are ignored from the sequence.The warning is displayed if the user submit input from both sources or no sequence is submitted.

Model Selection

There are three different models to select:
1. LBtope_Fixed is 20 mer model developed on LBtope_Fixed dataset.A fixed length of 20 window length pattern will be generated and predicted using the LBtope_Fixed model.

2. LBtope_Variable model is developed on variable length epitopes data (LBtope_Variable dataset. Since, model is developed on variable length epitopes, it can predict epitope having any length. The default size of window length is 15 as per Kringelum et al, 2013.

3. LBtop_Confirm model is developed on LBtop_Confirm dataset. The model is developed on those epitope, which are verified at least by two studies. This model is also developed on variable length epitopes, it can predict epitope having any length. The default size of window length is 15 as per Kringelum et al, 2013.

4. LBtop_Fixed_non_redundant model is developed on LBtop_Fixed_non_redundant dataset. A fixed length of 20 window length pattern will be generated and predicted using the LBtope_Fixed model.

5. LBtope_Variable_non_redundant model is developed on variable length epitopes data (LBtope_Variable_non_redundant dataset. Since, model is developed on variable length epitopes, it can predict epitope having any length. The default size of window length is 15 as per Kringelum et al, 2013.


E-mail Address
 
This is an optional field. Depending on the server load and number of sequences and length of each sequence, LBtope might take some time to serve the query. User can provide the Email ID to get the result through email however result will be printed on the screen itself after job completion. The page can be bookmarked for later reference.

% Probability Threshold and Output

During prediction each overalpping 20 mers of the sequence gets a SVM score. The scores are scaled from 20-100%. User need to set the % probability threshold above which epitopes should be printed. By Default it is 60%. If you want more confidence or high specificity in the prediction, we recommend setting higher threshold value. High scoring patterns will be bigger in fonts and epitopes will be printed in decreasing order of scores.

Result page output
Output from Sequence submission
Result page of a sequence will show a colourfull block for each sequence. Colour of amino acid is according to percent probability. On mouseover an amino acid, a popup window with results will show up, as shown below.







Output from Peptide submission and Mutant generation


Go to Top

Multiple peptides

LBtope server discriminate the linear B cell epitopes and non-epitopes for a given set of multiple peptides of various length. User can submit multiple peptides in FASTA format. The LBTOPE server uses the SVM models using dipeptide composition generated from the query sequence(s). The overall accuracy of this server is ~ 81%.

Sequence Input and Format

Amino acid sequences must be entered in the standard FASTA format. Sequence(s) can be pasted or can be given as a file. All the non standard characters like [*&^%$@#!()_+~=;'",<>?.\|} are ignored from the sequence.The warning is displayed if the user submit input from both sources or no sequence is submitted.

Model Selection

There are two different models to select:
1. LBtope_Variable model is developed on variable length epitopes data (LBtope_Variable dataset. Since, model is developed on variable length epitopes, it can predict epitope having any length. The default size of window length is 15 as per Kringelum et al, 2013.

2. LBtop_Confirm model is developed on LBtop_Confirm dataset. The model is developed on those epitope, which are verified at least by two studies. This model is also developed on variable length epitopes, it can predict epitope having any length. The default size of window length is 15 as per Kringelum et al, 2013.

3. LBtope_Variable_non_redundant model is developed on variable length epitopes data (LBtope_Variable_non_redundant dataset. Since, model is developed on variable length epitopes, it can predict epitope having any length. The default size of window length is 15 as per Kringelum et al, 2013.


E-mail Address
 
This is an optional field. Depending on the server load and number of sequences and length of each sequence, LBtope might take some time to serve the query. User can provide the Email ID to get the result through email however result will be printed on the screen itself after job completion. The page can be bookmarked for later reference.

% Probability Threshold and Output

During prediction each overalpping 20 mers of the sequence gets a SVM score. The scores are scaled from 20-100%. User need to set the % probability threshold above which epitopes should be printed. By Default it is 60%. If you want more confidence or high specificity in the prediction, we recommend setting higher threshold value. High scoring patterns will be bigger in fonts and epitopes will be printed in decreasing order of scores.

Result page output
Output from Sequence submission
Result page of a sequence will show a table having the peptides with their percent probability score along with physical properties score, as shown below.




Go to Top

Peptide mutatants

LBtope server generate all possible mutants of a peptide. It also predict B-cell epitope in these peptide mutants. This subroutine is important for identification of minimum mutation in a peptide in order to make it strong/weak B-cell epitope. The LBTOPE server uses the SVM models using dipeptide composition generated from the query sequence(s). The overall accuracy of this server is ~ 81%.

Sequence Input and Format

Amino acid sequence must be entered in the standard PLAIN format. Sequence(s) can be pasted. All the non standard characters like [*&^%$@#!()_+~=;'",<>?.\|} are ignored from the sequence.

Model Selection

There are two different models to select:
1. LBtope_Variable model is developed on variable length epitopes data (LBtope_Variable dataset. Since, model is developed on variable length epitopes, it can predict epitope having any length. The default size of window length is 15 as per Kringelum et al, 2013.

2. LBtop_Confirm model is developed on LBtop_Confirm dataset. The model is developed on those epitope, which are verified at least by two studies. This model is also developed on variable length epitopes, it can predict epitope having any length. The default size of window length is 15 as per Kringelum et al, 2013.

3. LBtope_Variable_non_redundant model is developed on variable length epitopes data (LBtope_Variable_non_redundant dataset. Since, model is developed on variable length epitopes, it can predict epitope having any length. The default size of window length is 15 as per Kringelum et al, 2013.


E-mail Address
 
This is an optional field. Depending on the server load and number of sequences and length of each sequence, LBtope might take some time to serve the query. User can provide the Email ID to get the result through email however result will be printed on the screen itself after job completion. The page can be bookmarked for later reference.

Result page output
Output from Sequence submission
Result page of a sequence will show a table having the original peptides with its percent probability score along with physical properties score. Next table shows all the possible mutant of the original peptide with their percent probability score along with physical properties score, as shown below




Go to Top