Figure 2: Identification of genomic fragments derived from distinct populations.
From: Community-integrated omics links dominance of a microbial generalist to fine-tuned resource usage
(a) Binning of assembled contigs (≥1,000 bp in length) on the basis of pentanucleotide signatures and visualization using the BH-SNE algorithm followed by human-augmented clustering of composite genome (CG) groups. (b) Violin plots of the G+C percentage for contigs within each of the CG groups. (c) Percentage amino-acid identity of the two subpopulations in CG8 (CG8a and CG8b) compared with the two sequenced Candidatus Microthrix parvicella (Bio17-1 (ref. 16) and RN1 (ref. 17)) genomes. The values are median±s.d. and n is the number of putative orthologues identified as best BLAST hits. Boxplots represent the lower quartile, median and upper quartile. Whiskers are placed at × 1.5 interquartile range beyond the lower and upper quartiles.
