Expanding Bioactive Fragment Space with the Generated Database GDB-13s

Abstract

Identifying innovative fragments for drug design can help medicinal chemistry address new targets and overcome the limitations of the classical molecular series. By deconstructing molecules into ring fragments (RFs, consisting of ring atoms plus ring-adjacent atoms) and acyclic fragments (AFs, consisting of only acyclic atoms), we find that public databases of molecules (i.e., ZINC and PubChem) and natural products (i.e., COCONUT) contain mostly RFs and AFs of up to 13 atoms. We also find that many RFs and AFs are enriched in bioactive vs inactive compounds from ChEMBL. We then analyze the generated database GDB-13s, which enumerates 99 million possible molecules of up to 13 atoms, for RFs and AFs resembling ChEMBL bioactive RFs and AFs. This analysis reveals a large number of novel RFs and AFs that are structurally simple, have favorable synthetic accessibility scores, and represent opportunities for synthetic chemistry to contribute to drug innovation in the context of fragment-based drug discovery.

One chiral fingerprint to find them all

“…We tested the ability of our chiral fingerprints for this task for small and large molecules separately. As a test case for small molecules, we computed Jaccard distances between all pairs involving the 203 structural isomers of 1,4-diaminocyclohexane ( 15 ), a ring fragment which is enriched in bioactive molecules from ChEMBL [ 57 , 58 ], and between all pairs of stereoisomers in the set. We similarly analyzed all pairs involving the 48 structural isomers of 4-aminopiperazine ( 16 ), a similar drug scaffold, and the stereoisomeric pairs within the set.…”

One chiral fingerprint to find them all

“…We tested the ability of our chiral fingerprints for this task for small and large molecules separately. As a test case for small molecules, we computed Jaccard distances between all pairs involving the 203 structural isomers of 1,4-diaminocyclohexane (15), a ring fragment which is enriched in bioactive molecules from ChEMBL, 57,58 and between all pairs of stereoisomers in the set. We similarly analyzed all pairs involving the 48 structural isomers of 4-aminopiperazine ( 16), a similar drug scaffold, and the stereoisomeric pairs within the set.…”

Navigating a 1E+60 Chemical Space

“…Since the advent of combinatorial chemistry in the early 1990's, which was triggered by the invention of the split-and-mix method yielding one-bead-one-compound libraries of millions of peptide and peptide-like oligomers in a few tens of synthetic operations, [1][2][3] drug discovery has been fascinated and partly driven by large numbers. [4][5][6] Approaches ranged from the "needle in a haystack" method of high-throughput screening typical for genetically encoded display libraries 7,8 and DNA-encoded libraries, 9,10 to the concept of chemical space guiding the design of focused libraries of small druglike molecules, [11][12][13] fragments 14,15 and peptides. [16][17][18] Many projects are currently exploiting "makeon-demand" virtual libraries of a few billion members obtained by using various coupling chemistries to combine two to four building blocks, each being taken from a pool of thousands of building blocks, to form linear, branched or cyclic oligomers.…”