BEST: B-cell Epitope prediction using Support vector machine Tool

This web page provides datasets and prediction model associated with:
Gao J, Faraggi E, Zhou Y, Ruan J, Kurgan L. BEST: improved prediction of B-cell epitopes from antigen sequences. PLoS One. 2012; 7(6):e40104
  1. Datasets
    Each dataset is provided in .txt format.

      ChenFrag dataset is available at this link
      BCPREDFrag dataset is available at this link
      Filtered40_BCPREDFrag
          the training dataset with epitopic 20-mers can be downloaded from here: Filtered40_BCPREDFrag.pos.txt
          the training dataset with non-epitopic 20-mers can be downloaded from here: Filtered40_BCPREDFrag.neg.txt
      The SEQ197 test dataset can be downloaded from here: SEQ194.txt and its source is available at this link
          each proteins in this dataset is represented by UniProt ID, sequence, and annotation of epitopic residues (using 1) and non-epitopic residues (using 0)
      The SEQ19 test dataset can be downloaded from this link

  2. Prediction model
    The model is packaged using ZIP format and can be downloaded from here: BEST model ver 1.1.
    The ZIP file includes instructions (in README.txt file) how to prepare and use the predictor.