Table 7 Table of true positive rate, false positive rate, L2 loss and prediction error for the additive-log-ratio model with SNR of 2.5 and d of 45
From: Bayesian compositional regression with microbiome features via variational inference
Method | \(\rho\) | TPR | FPR | L2 | PE |
|---|---|---|---|---|---|
Lasso | 0 | 1 ± 0 | 0.397 ± 0.159 | 3.647 ± 0.386 | 0.211 ± 0.125 |
Lin | 0 | 1 ± 0 | 0.374 ± 0.144 | 3.668 ± 0.377 | 0.214 ± 0.122 |
Bates | 0 | 1 ± 0 | 0.022 ± 0.034 | 3.914 ± 0.303 | 0.191 ± 0.119 |
Selbal | 0 | 0.790 ± 0.073 | 0.000 ± 0.000 | 8.645 ± 0.562 | 1.637 ± 0.526 |
VB | 0 | 1 ± 0 | 0.015 ± 0.019 | 3.848 ± 0.278 | 0.197 ± 0.119 |
Lasso | 0.2 | 1 ± 0 | 0.297 ± 0.134 | 3.723 ± 0.360 | 0.184 ± 0.107 |
Lin | 0.2 | 1 ± 0 | 0.279 ± 0.137 | 3.817 ± 0.366 | 0.196 ± 0.113 |
Bates | 0.2 | 1 ± 0 | 0.019 ± 0.033 | 3.957 ± 0.284 | 0.169 ± 0.102 |
Selbal | 0.2 | 0.667 ± 0 | 0 ± 0 | 9.161 ± 0.404 | 1.182 ± 0.369 |
VB | 0.2 | 1 ± 0 | 0.004 ± 0.010 | 3.870 ± 0.267 | 0.159 ± 0.091 |
Lasso | 0.4 | 1 ± 0 | 0.366 ± 0.145 | 3.544 ± 0.371 | 0.226 ± 0.163 |
Lin | 0.4 | 1 ± 0 | 0.429 ± 0.139 | 3.533 ± 0.393 | 0.229 ± 0.150 |
Bates | 0.4 | 1 ± 0 | 0.019 ± 0.035 | 3.811 ± 0.316 | 0.186 ± 0.115 |
Selbal | 0.4 | 0.667 ± 0 | 0 ± 0 | 8.192 ± 0.386 | 0.707 ± 0.279 |
VB | 0.4 | 1 ± 0 | 0.017 ± 0.017 | 4.104 ± 0.313 | 0.208 ± 0.126 |