Figure 1.
Flowchart of the ADE detection process for pancreatitis.
Figure 2.
Receiver Operating Characteristic (ROC) (a) and Precision-Recall (b) curves evaluating the set of 278 EHR ADE candidates with OR05 and different MFBMs.
It is worth noting that although OR05 algorithm is very useful to originate the first set of 278 candidate drugs related to pancreatitis (99 out of 278 drugs were already included in the pancreatitis reference standard set), the precision of the method is constant within this set. However, an improvement of the precision in top positions can be achieved using MFBM (in the graphic: black-OR05, red-MACCS, green-GpiDAPH3, yellow-TGT, blue-TGD).
Table 1.
Performance of DPA compared to DPA+MFBM (DPA combined with MACCS, GpiDAPH3, TGD and TGT molecular fingerprints) in different TOP positions.
Figure 3.
Receiver Operating Characteristic (ROC) (a) and Precision-Recall (b) curves evaluating the test set of EHR pancreatitis candidates (in the graphic are not included the drugs already in the reference standard: 21 true positives versus 158 false positives, black-OR05, red-MACCS, green-GpiDAPH3, yellow-TGT, blue-TGD).
Table 2.
Examples of candidates selected through the combination of DPA and MFBM (MACCS fingerprints) and similar molecules in the pancreatitis reference standard, along with OR05 (lower 5th percentile of the Odds Ratio measure of association in DPA analysis) and TC (Tanimoto coefficient) values.