Fig. 1
From: Strain-level metagenomic assignment and compositional estimation for long reads with MetaMaps
Read assignment accuracy in experiments with simulated data (i100, p25, and CAMI). Bar plots show precision, recall (read level), and recall (base level) for MetaMaps, Kraken, Kraken 2, Centrifuge and LAST + MEGAN-LR at different evaluation levels. Note that Kraken, Kraken 2 and MEGAN-LR were not designed to achieve strain-level resolution and therefore these tools are not evaluated at this level; also note that the CAMI experiment is not evaluated at the strain level because the CAMI truth set data are not consistently strain-level
