Fig. 5From: Molecular complex detection in protein interaction networks through reinforcement learningA synthetic disconnected toy network of complexes. Complexes A–G are used for training and H–N are used for evaluating the RL algorithmBack to article page