A View on Molecular Complexity from the GDB Chemical Space

One recurring question when choosing which molecules to select for investigation is that of molecular complexity: is there a price to pay for complexity in terms of synthesis difficulty, and does complexity have anything to do with biological properties? In the chemical space of small organic molecules enumerated from mathematical graphs in the GDBs (Generated DataBases), most compounds are too complex and challenging for synthesis despite containing only standard functional groups and ring types. For these GDB molecules, we find that an increasing fraction (MC1) or number (MC2) of non-divalent nodes in the molecular graph represent simple measures of molecular complexity, which we interpret in terms of potential synthesis difficulties. We also show that MC1 and MC2 are applicable to commercial screening compounds (ZINC), bioactive molecules (ChEMBL) and natural products (COCONUT) and compare them with previously reported measures of molecular complexity and synthetic accessibility.