查看文章

Supplementary Table SF1. T1 summarizes the number of entries for each non-redundant set of protein-small molecule and proteinprotein complexes. Some classifications inevitably exhibit a certain degree of ambiguity, for example the definition of “natural product” is inherently rather arbitrary. By the same argument,“drug-like molecule” classification is not unequivocal; it is more a continuous “likeness” property without rigorous thresholds. Furthermore, the emerging new targets have forced debate about what it takes to be a drug [1]. The molecular property thresholds applied to select “drug-like” molecules in this study are somewhat loose, for instance the molecular weight cut-off is 900. The reason for these broad filters is to be able to compare like to like with the small molecules inhibiting protein-protein complexes (molecular weight range: 150-815Da). Therefore, the term “synthetic small molecules” is used instead of drug-like across the manuscript. In other cases, the annotation seems to be accurate and straightforward but