Fast and space-efficient taxonomic classification of long reads with hierarchical interleaved XOR filters

Average scores on species level across the three long-read mock community data sets. (A) Average precision and recall of the six compared tools on the species level for the two ONT and the HiFi mock community data sets. MetaMaps, Taxor, and Ganon have the highest average recall, whereas Taxor and KMCP outperform the other tools in terms of precision. KMCP is the only tool with an average recall below 0.5 across the three data sets. (B) Average species-level F₁-score of the six tools across the three real mock communities. Taxor and MetaMaps show the highest scores, whereas KMCP has the lowest average F₁-score. (C) Average species-level F_0.5-score of the six tools across the three real mock communities. Our new tool, Taxor, outperforms the other methods when precision is prioritized over recall.