“…Most of those models are indeed not compliant with the OECD guidance on QSAR model development and validation [82]. More specifically, the primary drawbacks of the majority of published QSAR studies are: (i) most models do not have proof of passing the Y-randomization test [21, 23, 26, 28, 29, 31–35, 38, 40, 41, 45–49, 51–56, 58, 59, 63–65, 68–70, 75, 79]; (ii) no proof of applicability domain (AD) estimation is provided [21, 23, 26–29, 31–36, 40, 45, 48–53, 56, 58, 63–65, 68–71, 75, 79]; and (iii) model predictivity is not acceptable [39, 61, 66]. As a consequence, despite the large number of QSAR models for hERG blockage available in the literature, only very few models can actually be employed to predict hERG blockage [60, 61, 74, 78].…”