The reviewer highlights a key reason we focus on the R2 and Spearman metrics for this work - such scaling differences are removed in both metrics. Moreover, the ANI developers noted a difference between their published version of the ANI-2x model and the pre-release version we used initially. For example, in the initial manuscript, molecules containing chlorine were not included.
In the revised manuscript, we have revised the ANI-2x data, fixing conversion factors as well as using the final ANI-2x model on all available molecules. Very little changed in the tables, figures, or conclusions.