Fig. 3. Features of the precalculated search space of 30 000 6-mer molecules; (a) different representative building blocks for 6 different clusters of oligomers; (b) Violin representation of the distribution of Fcomb values of the oligomers with dominant building blocks from different clusters. Here we defined the “dominant building block cluster” as the most frequent building block in the molecule, and when there is no dominant building block cluster, we classify the constructed molecules as “mixed”; the dashed lines split the distribution into 4 quartiles. (c) The frequency of molecules in each cluster of dominant building block clusters.
