Fast prediction of distances between synthetic routes with deep learning

Abstract We expand the recent work on clustering of synthetic routes and train a deep learning model to predict distances between arbitrary routes. The is based long short-term memory representation route trained as twin network reproduce tree edit distance (TED) two machine approach approximately orders magnitude faster than TED enables many more from retrosynthesis prediction. clusters have high degree similarity given by TED-based are accordingly intuitive explainable. provide developed open-source.