Skip to main content
. 2011 Dec 14;6(12):e27310. doi: 10.1371/journal.pone.0027310

Table 1. Ability of chimera.slayer, Uchime, and chimera.perseus to detect reliably chimeras within the sliding window and shhh.flows pipelines when using the Gold sequence collection or the dataset itself as the reference.

Sequence curation pipeline Chimera detection algorithm Database Sensitivitya , b Specificity False discovery rate Initial % chimeras Final % chimeras
Sliding window chimera.slayer Gold 68.7 (11.8) 95.0 (3.4) 3.7 (3.4) 67.8 (11.9) 39.6 (8.4)
Self 66.5 (11.8) 94.2 (3.7) 4.4 (4.1) 67.8 (11.9) 41.4 (9.3)
Uchime Gold 79.2 (8.0) 93.7 (3.7) 4.1 (3.3) 67.8 (11.9) 30.6 (8.7)
Self 74.5 (10.5) 94.4 (3.7) 4.0 (3.7) 67.8 (11.9) 34.6 (8.5)
chimera.perseus Self 76.0 (5.9) 91.6 (5.0) 5.6 (4.9) 67.8 (11.9) 34.8 (10.6)
shhh.flows chimera.slayer Gold 79.7 (7.7) 95.9 (2.6) 1.9 (1.2) 71.8 (7.3) 33.8 (13.2)
Self 79.5 (7.1) 94.4 (3.6) 2.5 (1.4) 71.8 (7.3) 34.2 (12.3)
Uchime Gold 89.6 (6.3) 93.7 (3.9) 2.6 (1.7) 71.8 (7.3) 20.8 (11.3)
Self 87.3 (5.8) 93.8 (4.0) 2.6 (1.6) 71.8 (7.3) 24.5 (10.8)
chimera.perseus Self 87.4 (6.8) 92.7 (4.3) 2.9 (1.5) 71.8 (7.3) 24.4 (12.4)
a

All values in this table were calculated using unique sequences.

b

The standard deviation is given in parentheses.