Skip to main content
. Author manuscript; available in PMC: 2021 Feb 24.
Published in final edited form as: IEEE/ACM Trans Comput Biol Bioinform. 2018 May 29;16(3):1029–1035. doi: 10.1109/TCBB.2018.2841396

Fig. 7.

Fig. 7.

Receiver operating characteristic curves for a binarized version of this classification problem (no-hit versus soft-hit or hard-hit). ROC-AUC scores range between 0.5 and 1.0, with 0.5 indicating performance at chance and 1.0 indicating perfect classification (a standard which the best custom convolutional neural network we tried achieves on this binarized problem).