Welcome to Pfeature

Pfeature is a comprehensive web server which will allow users to compute most of the protein features that have been discovered over the past decades. Using different functional modules of this web service, users will be able to evaluate four major categories of protein features: i) Composition-based features, ii) Binary profile of sequences, iii) evolutionary information based features, and iv) structural descriptors, for a group of protein/peptide sequences. Additionally, users will also be able to generate these features for sub-parts of protein/peptide sequences. This will be helpful to annotate structure, function and therapeutic properties of proteins.

Major features

Composition based

This Pfeature service consists of functional modules for evaluation of various composition based features such as amino acid composition (residue level, dipeptide level, tripeptide level) and atomic composition. Features based on physico-chemical properties of amino acids, residue repeats and distribution, distance distribution of residues and shannon information content can also be computed by the user here. Other functionalities include descriptors such as autocorrelation, conjoint triad, composition transition, pseudo amino acid composition and quasi-sequence-order descriptors. Users will be able to calculate most of these features for both protein sequences and their sub-sequences.

Binary profiles based

Binary profiles of protein sequences for amino acid, dipeptide, physico-chemical properties and atomic composition can be evaluated using this module. These features can be also be evaluated for sub-sequences, as desired by the user.

Evolutionary Information

Modules under this category generate PSSM (Position Specific Scoring Matrix) based features like PSSM compostion, PSSM Profiles, PSSM matrix. PSSM profiles provides information of the residue conservation in the given protein sequence. This kind of features have been heavily used in developing different types of structures and turns prediction methods, protein residue annotation prediction

Structure based

Modules under this category generate structure based features includes 2D and 3D descriptors, different types of Fingerprints. Also the module generates features in the form of SMILES format using tertiary structure.This kind of features have been heavily used in developing different types of prediction methods and drug designing methods.