Training, validation and testing sample data used in "Antibody interface prediction with 3D Zernike descriptors and SVM"

This dataset contains the generated training, validation and testing samples used in the paper "Antibody interface prediction with 3D Zernike descriptors and SVM".<br>3D Zernike descriptors are used to represent protein surface shape, while Support-Vector Machine (SVM) is the binary classification technique employed to produce a model based on the training data which predicts the class labels of the test data given only the feature vectors of the test data.<br>Data are archived in the format .tar.xz, which can be extracted by common archive utilities. The archive contains a number of text files in the SVM light format. Each line records 1331 colon-separated pairs of numbers (the first one being a feature index - an integer ranging from 1 to 1331, the second a floating point number). <br>Background:In the related paper we present a novel method for antibody interface prediction from their experimentally-solved structures based on 3D Zernike Descriptors. Roto-translationally invariant descriptors are computed from circular patches of the antibody surface enriched with the physicochemical properties from the HQI8 amino acid index set, and are used as samples for a binary classification problem. A SVM classifier is used to distinguish interface surface patches from non-interface ones. By exploiting the spatial continuity of the antigen-binding regions, the Isolation Forest algorithm is used to discard false-positive patches isolated from the others. Each residue is assigned a score by the overlying predicted patches indicating its likelihood of belonging to the binding region: a residue is identified as belonging to the interface only if its score reaches a minimum threshold value. The proposed method was shown to outperform other state-of-the-art antigen-binding interface prediction software.