Skip to main content

Table 2 The eHOMD training set is superior for assigning species/supraspecies-level taxonomy to short- and long-read human aerodigestive tract datasets

From: Construction of habitat-specific training sets to achieve species-level assignment in 16S rRNA gene datasets

  V1V3_hADT_CL V1V3_HMPnares_ ASVb FL_sinonasal_SMRT_ASV
(% reads) (% ASVs) (% reads) (% ASVs) (% reads)
eHOMD Genus 100.0 95.5 98.9 99.5 100.0
Species 100.0 93.9 98.5 95.1 99.0
SILVA Genus 96.1 94.7 97.6 96.6 98.9
Species 44.7a 4.1a 29.9a 18.6a 71.9a
RDP Genus 93.2 90.2 92.2 94.1 98.5
Species 38.5a 3.1a 27.5a 13.2a 60.6a
  1. aExact match algorithm
  2. bASVs derived from the HMP nares V1–V3 dataset, as described in [20], constitute the V1V3_HMPnares_ASV dataset (Additional file 10)