We have previously reported on classes of compounds that can interfere with bioassays via a number of different mechanisms and termed such compounds Pan Assay INterference compoundS, or PAINS. These compounds were defined on the basis of high-throughput data derived from vendor-supplied synthetics. The question therefore arises whether the concept of PAINS is relevant to compounds of natural origin. Here, it is shown that this is indeed the case, but that the context of the biological readout is an important factor that must be brought into consideration.