Skip to main content
. Author manuscript; available in PMC: 2019 May 15.
Published in final edited form as: Science. 2018 Oct 19;362(6412):347–350. doi: 10.1126/science.aat4807

Fig. 2. Prediction of drug-target interactions.

Fig. 2.

(A) Predictions from the DrugBank 3.0 dataset. Bar height corresponds to mean AUPR (area under the precision-recall curve) and error bars correspond to the standard deviation. We compared Secure DTI to the plaintext methods BLMNII (20), NetLapRLS (16), HNM (21), MF (13), CMF (14), and DTINet (15) as reported in Luo et al. (15) by means of 10-fold cross validation (CV) on balanced training and test sets (Methods; see Fig. S1 for other evaluation settings). (B) Predictions form the STITCH 5 dataset with over a million drug-target pairs. Secure DTI is compared to matrix factorization with (CMF) and without (MF) side information (see Fig. S2 for other evaluation settings). Solid line, sampling negative examples randomly; and dashed line, sampling negative examples while matching the relative frequencies of drugs and targets to those in the positive examples, representing a more challenging test case. Reported AUPRs are for the solid curves. (C) Runtime of our training protocol, over a local area network (LAN), for different dataset sizes. Box height represents the standard deviation.