Computational Bioactivity Fingerprint Similarities To Navigate the Discovery of Novel Scaffolds

As
one of the central tasks of modern medicinal chemistry, scaffold
hopping is expected to lead to the discovery of structural novel biological
active compounds and broaden the chemical space of known active compounds.
Here, we report the computational bioactivity fingerprint (CBFP) for
easier scaffold hopping, where the predicted activities in multiple
quantitative structure–activity relationship models are integrated
to characterize the biological space of a molecule. In retrospective
benchmarks, the CBFP representation shows outstanding scaffold hopping
potential relative to other chemical descriptors. In the prospective
validation for the discovery of novel inhibitors of poly [ADP-ribose]
polymerase 1, 35 predicted compounds with diverse structures are tested,
25 of which show detectable growth-inhibitory activity; beyond this,
the most potent (compound 6) has an IC50 of 0.263 nM. These
results support the use of CBFP representation as the bioactivity
proxy of molecules to explore uncharted chemical space and discover
novel compounds.