Skip to main content
Fig. 5 | Biology Direct

Fig. 5

From: MEGAN-LR: new algorithms allow accurate binning and easy interactive exploration of metagenomic long reads and contigs

Fig. 5

Violin plots comparing the performance of LAST+MEGAN-LR and Kaiju for two simulation studies, one based on a R7.3 Nanopore chemistry profile and the other based on a R9 Nanopore chemistry profile. In both cases, we report the sensitivity (percentage of reads assigned to the correct taxon) and precision (percentage of reads assigned correctly out of all reads not binned to an ancestor of the correct taxon) of taxonomic assignments. This is done at the genus level for nine different categories of genera (reflecting the number of species in the genus from which the target species was removed), and for all. Results for the R7.3 profile are shown in a and b, and results for the R9 profile are shown in c and d

Back to article page