PB-SVindex help
PB-SVindex
PB-SVindex : knowledge-based calculation of Structural Variability Indices of pentapeptide sequences
This tool uses the knowledge of local conformations of pentapeptides in terms of PBs from PB-PENTAdb database to calculate a structural variability index (SVI) for each of them. It shows if a pentapeptide is observed to have multiple local conformations in terms of PBs or not.
This proposed index is similar to the Simpson’s diversity index used in population genetics :
pi is the fraction of all organisms which belong to the i-th species
Here we calculate SVI using an approximation of Simpson’s index :
where, for a given amino acid motif, ni is its occurence in PB class i, and n is the total number of times this motif is found in the database.
This is an estimator for Simpson’s index for sampling without replacement.
To avoid any bias in the calculation of this index, we used a non-redundant dataset (Astral40). However, in such a dataset, we cannot calculate SVI values for many pentapeptides because of their low occurrence. We hence proposed to calculate the SVI values based on the two constitutive tetrapeptides of each pentapeptide : this SVI is an average of SVI values calculated for the first four and last four residues (see figure below).
Assigned variability index values vary between 0 and 1. Lower values indicate sequences with low structural variability, higher values indicate higher structural variability.
User can provide a SCOP domain ID to get the SVindex values for its sequence.
The output of this query will give a result page as follows
User can choose to provide one more protein sequences in fasta format (1000 sequences at most) as shown below
As output, a table is provided with
- first 8 characters of ID of sequence (it removes the “>” from first line)
- Number of residues in sequence
- SV index
with SVindices values separated by comas.
