Data Collection

Antioxidant proteins together with their information including taxonomy, source organism, subcellular location, gene ontology, catalytic activity and function were obtained from the UniProtKB/Swiss-Prot database (release 2016_11) by searching the keyword “antioxidant”. In order to obtain the quality data, the following steps were performed: (i) only proteins with the experimentally confirmed antioxidative activities were included; (ii) proteins including illegal letters, i.e., “B”, “X” or“Z”, were excluded.This finally yields 710 proteins that have experimentally proven or confirmed antioxidative activity to be included in the current database.

Search Options

Users can basically search antioxidant proteins by entering the related query items in the searching fields, such as UniProt ID, Taxonomy, Organism, Subcellular locations or Protein names. To perform conditional search, users can add (or remove) the searching fields using the Add (or Remove) button and then join the multiple query items by logical operators AND/OR.



AodPred is a support vector machine (SVM) based model and can be used to predict whether a protein is antioxidant or not. Users can paste or type sequence/sequences in fasta format to make predictions. The probability (0 to 1) of the predictions being antioxidant proteins or non-antioxidant proteins will be displayed.


This module allows users to perform similarity based search against the experimentally confirmed antioxidant proteins in AOD using the BLAST program. After submission of the query protein sequences, a list of antioxidant proteins similar to the query protein sequences will be returned.


This interface assists the users understand the compositions of the AOD. It provides the information about organism distribution of antioxidant proteins, length distribution of antioxidant proteins, subcellular location distribution of antioxidant proteins and relative amino acid composition of antioxidant proteins in AOD.