Fig. 8: The EPI2ME analysis platform yields inaccurate genus-level data.
From: Cross-comparison of gut metagenomic profiling strategies
Results from the 16S module of the EPI2ME Labs program by ONT employed on canine stool sample (a,b), fecal samples from six additional dogs (c), neonatal54 (d) and adult56 (e) human stool specimens, adult human skin specimen57 (f), and MCS (g). In our canine samples, the software failed to yield abundances to the Peptacetobacter, Faecalimonas, and Mediterraneibacter genera, discarding 75% of the reads and skewing the actual compositional ratios. * In (b, c, d, e, f, and g) of this figure the blocks highlighted in red represent the reads that the EPI2ME program filters out and excludes from the analysis due to the LCA tag.
