Skip to main content

Table 2 Sensitivity and specificity of filtering software over different error types

From: Evaluating the usefulness of alignment filtering methods to reduce the impact of errors on evolutionary inferences

sensitivity prokaryotic MSAs eukaryotic MSAs
  frameshifts scrambled insertions frameshifts scrambled insertions
HmmCleaner 96.95% 96.24% 99.24% 95.99% 96.26% 99.34%
PREQUAL 93.13% 96.77% 99.83% 85.64% 97.34% 99.95%
BMGE 16.00% 19.23% 97.55% 8.57% 12.57% 96.16%
OD-seq 20.11% 25.09% 59.89% 8.08% 9.34% 59.80%
GUIDANCE2 4.90% 5.37% 0.25% 2.33% 3.25% 1.69%
specificity prokaryotic MSAs eukaryotic MSAs
  frameshifts scrambled insertions frameshifts scrambled insertions
HmmCleaner 87.28% 87.24% 85.64% 94.70% 94.70% 94.63%
PREQUAL 92.80% 92.78% 92.76% 94.77% 94.65% 94.59%
BMGE 91.01% 90.88% 91.10% 97.05% 96.93% 96.94%
OD-seq 94.48% 94.46% 92.55% 95.20% 95.25% 94.90%
GUIDANCE2 99.55% 99.48% 99.85% 98.47% 98.35% 98.51%