Fig. 2From: Predicting protein network topology clusters from chemical structure using deep learningData pre-processing for CNN and RNN architectures. A An example of SMILES string conversion to a matrix of dimension 42 X (length of SMILES string). This matrix is padded along the y-axis up to a defined maximum length. B (a) Atomwise and (b) SMILES-PE tokenization for the compound aspirinBack to article page