From: A statistical framework for detecting mislabeled and contaminated samples using shallow-depth sequence data
Method
L = 50
L = 100
L = 250
L = 500
L = 1000
BIGRED
0.9832
0.9895
0.9958
0.9973
0.9981
Hierarchical clustering
0.8322
0.9088
0.9488
0.9640
0.9771