Skip to main content
. 2021 Mar 13;10(3):219. doi: 10.3390/biology10030219

Table A1.

Combinations of top prediction models and their diagnostic performance in Train and CV set.

  • (A)

    Phylum (L2) level

Number of Variables Model Train Set Cross-Validation (CV) Set
Sensitivity Specificity AUC Sensitivity Specificity AUC
0 class~age + sex 0.311 0.846 0.598 0.206 0.846 0.531
1 class~age + sex + p16 0.944 0.846 0.953 0.839 0.846 0.886
2 class~age + sex + p16 + p6 1.000 0.846 0.966 1.000 0.846 0.962
3 class~age + sex + p16 + p6 + p2 1.000 0.846 1.000 0.783 0.885 0.863
4 class~age + sex + p16 + p6 + p2 + p11 1.000 0.846 1.000 0.833 0.885 0.897
5 class~age + sex + p16 + p6 + p2 + p11 + p19 1.000 0.846 1.000 0.733 0.846 0.855
6 class~age + sex + p16 + p6 + p2 + p11 + p19 + p18 1.000 0.846 1.000 0.400 0.923 0.844
7 class~age + sex + p16 + p20 + p6 + p2 + p11 + p19 + p18 1.000 0.846 1.000 0.400 0.923 0.867
8 class~age + sex + p16 + p20 + p6 + p2 + p13 + p11 + p19 + p18 1.000 0.846 1.000 0.578 0.846 0.833
9 class~age + sex + p16 + p20 + p6 + p2 + p13 + p11 + p15 + p19 + p18 1.000 0.846 1.000 0.200 0.923 0.797
10 class~age + sex + p16 + p20 + p6 + p9 + p2 + p13 + p11 + p15 + p19 + p18 1.000 0.846 1.000 0.222 0.923 0.616
11 class~age + sex + p16 + p20 + p6 + p9 + p22 + p2 + p13 + p11 + p15 + p19 + p18 1.000 0.846 1.000 0.100 0.923 0.658
12 class~age + sex + p16 + p20 + p6 + p9 + p22 + p10 + p2 + p13 + p11 + p15 + p19 + p18 1.000 0.846 1.000 0.389 0.923 0.684
13 class~age + sex + p16 + p20 + p6 + p9 + p22 + p10 + p2 + p14 + p13 + p11 + p15 + p19 + p18 1.000 0.846 1.000 0.000 1.000 0.654
  • (B) 

    Genus (L6) level

Number of Variables Model Train Set Cross-Validation (CV) Set
Sensitivity Specificity AUC Sensitivity Specificity AUC
0 class~age + sex 0.3111 0.8462 0.5976 0.2056 0.8462 0.5314
1 class~age + sex + g150 0.9444 0.8462 0.9615 0.9444 0.8462 0.9299
2 class~age + sex + g150 + g64 1.0000 0.8462 1.0000 0.9500 0.9231 0.9515
3 class~age + sex + g150 + g64 + g22 1.0000 0.8462 1.0000 1.0000 0.9231 0.9957
4 class~age + sex + g15 + g150 + g64+g22 1.0000 0.8462 1.0000 0.9444 0.9231 0.9594
5 class~age + sex + g15 + g150 + g64 + g225 + g22 1.0000 0.8462 1.0000 0.9444 0.9231 0.9594
6 class~age + sex + g210 + g23 + g150 + g64 + g225 + g22 1.0000 0.8462 1.0000 1.0000 0.9231 0.9829
7 class~age + sex + g49 + g15 + g210 + g150 + g64 + g225 + g22 1.0000 0.8462 1.0000 1.0000 0.9231 1.0000
8 class~age + sex + g19 + g49 + g15 + g210 + g150 + g64 + g225 + g22 1.0000 0.8462 1.0000 0.8444 0.8846 0.9034
9 class~age + sex + g19 + g49 + g15 + g210 + g23 + g150 + g64 + g225 + g22 1.0000 0.8462 1.0000 0.7944 0.8462 0.8613
10 class~age + sex + g19 + g49 + g15 + g210 + g23 + g150 + g59 + g64 + g225 + g22 1.0000 0.8462 1.0000 0.3500 0.9231 0.8338

The finally selected model was highlighted as bold. The OTU symbols are as follows: p16, Verrucomicrobia; p6, Actinobacteria; g49, Sphingomonas; g15, Ruminococcaceae UCG-014; g210, Propionibacterium; g150, Akkermansia; g64, Ruminiclostridium; g225, Lachnospiraceae UCG-001; g22, Corynebacterium 1.