Skip to main content

Table 6 Feature selection performance using alignment-free methods (error, %)

From: Efficient alignment-free DNA barcode analytics

  # features selected
Dataset Full feature set (1048576 feat.) 4096 2048 1024 512 200 100
ACG 2.49 ± 0.87 2.51 ± 0.95 2.79 ± 1.02 3.00 ± 0.96 3.17 ± 0.86 3.52 ± 0.64 4.48 ± 0.86
Hesperiidae 3.57 ± 1.08 3.53 ± 1.12 3.80 ± 1.22 4.17 ± 1.05 4.40 ± 1.15 4.81 ± 1.30 5.64 ± 1.20
Astraptes 1.07 ± 1.81 0.44 ± 0.92 0.44 ± 0.92 0.44 ± 0.92 0.44 ± 0.92 0.64 ± 1.03 1.49 ± 1.75
Bats of Guyana 1.63 ± 1.22 1.63 ± 1.22 1.63 ± 1.22 1.63 ± 1.22 1.63 ± 1.22 1.63 ± 1.22 1.63 ± 1.22
Birds 6.30 ± 1.80 6.45 ± 1.82 6.94 ± 2.08 7.13 ± 2.05 7.41 ± 1.77 9.10 ± 1.64 9.84 ± 1.99
Fish of Australia 5.50 ± 3.27 5.35 ± 3.36 5.35 ± 3.36 6.14 ± 3.50 6.80 ± 3.15 8.32 ± 2.75 9.51 ± 2.40
Fish larvae 2.86 0 0 0 0 0 0