Fig. 5From: Predicting protein network topology clusters from chemical structure using deep learningWorkflow used in this study. The data was obtained from STITCH and STRING databases and were processed using Quantmap followed by hierarchical clustering using several distance thresholds. For each distance threshold, a subset of 20 clusters was used to evaluate different deep learning architectures. Further, a dataset of interest was selected for training and functional assignment of clusters was carried out. The final trained model was later evaluated using well-known chemicalsBack to article page