- Tools were run on three mock community datasets (ONT_Zymo, PB_ATCC and PB_Zymo) subsampled to three different sizes to test their scalability. All datasets used in the main part of the paper were subsampled to 1Gbp (file size is about 2 GB) to make the testing viable even for slower tools. However, all three datasets were used in their original size, subsampled to half of the original dataset and subsampled to 1 Gbp. The table also shows the execution time for real gut microbiome datasets in their full size. Since Bracken, MEGAN-N, MEGAN-P and MetaMaps running times were prohibitively long on larger datasets, we omitted them from the analysis. Additionally, deSAMBA was unable to process some larger datasets completely. The table shows execution time in seconds
