Fig. 5

All complete genomes of Archaea and Bacteria in RefSeq. The uncompressed data set has a size of about 98.8 GiB. Ten million query reads of length \(250\,bp\) were simulated using the Mason simulator [19]. The parameters used for all tools (if applicable) were: canonical k-mers \(k=32\), no k-mer filtering, false-positive rate \(5\,\%\), 2 hash functions, 32 threads, query search threshold 0.7. a Query time for varying number of transcripts, the best time of three runs was used. b Index build time, including all preprocessing steps. c Peak in RAM usage during querying for varying number of transcripts. d Index size stored on disk. Numeric values can be found in Additional file 1