One Among Millions: The Chemical Space of Nucleic Acid-Like Molecules

Biology encodes hereditary information
in DNA and RNA, which are
finely tuned to their biological functions and modes of biological
production. The central role of nucleic acids in biological information
flow makes them key targets of pharmaceutical research. Indeed, other
nucleic acid-like polymers can play similar roles to natural nucleic
acids both in vivo and in vitro;
yet despite remarkable advances over the last few decades, much remains
unknown regarding which structures are compatible with molecular information
storage. Chemical space describes the structures and properties of
molecules that could exist within a given molecular formula or other
classification system. Using structure generation methods, we explore
nucleic acid analogues within the formula ranges BC3–7H5–15O2–4 and BC3–6H5–15N1–2O0–4, where B is a recognition element (e.g., a nucleobase). Other restrictions
included two obligatory points of attachment for inclusion into a
linear polymer and substructures predicting chemical stability. These
sets contain 86,007 (CHO) and 75,309 (CHNO) compositionally isomeric
structures, representing 706,568 CHO and 454,422 CHNO stereoisomers,
that diversely and densely occupy this space. These libraries point
toward there being large spaces of unexplored chemistry relevant to
pharmacology and biochemistry and efforts to understand the origins
of life.