Table 2.
The benchmarking results on the non-structural alignments of the CMfinder dataset ( β=10)
| Family | MPI | RNAalifold | PhyloRNAalifold |
RNAalifold with RIBOSUM |
PhyloRNAalifold with RIBOSUM |
PETFold |
|---|---|---|---|---|---|---|
| Cobalamin |
43.2 |
-0.001 |
-0.001 |
-0.002 |
-0.002 |
-0.002 |
| ctRNA_pGA1 |
66.5 |
0.865 |
0.889 |
0.979 |
0.979 |
0.936 |
| Entero_CRE |
81.7 |
0.912 |
0.912 |
0.916 |
0.916 |
1.000 |
| Entero_OriR |
87.5 |
0.694 |
0.830 |
0.965 |
0.965 |
0.910 |
| glmS |
55.2 |
0.445 |
0.566 |
0.873 |
0.784 |
0.811 |
| Histone3 |
45.1 |
1.000 |
1.000 |
1.000 |
1.000 |
1.000 |
| Intron_gpII |
46.2 |
0.760 |
0.794 |
0.826 |
0.826 |
0.794 |
| IRE |
77.3 |
0.480 |
0.480 |
0.710 |
0.710 |
0.816 |
| let-7 |
66.7 |
0.760 |
0.761 |
0.666 |
0.666 |
0.742 |
| lin-4 |
64.6 |
0.523 |
0.521 |
0.712 |
0.712 |
0.739 |
| Lysine |
44.0 |
0.307 |
0.414 |
0.513 |
0.484 |
0.388 |
| mir-10 |
68.4 |
0.741 |
0.820 |
0.957 |
0.957 |
0.935 |
| Purine |
53.8 |
0.852 |
0.977 |
0.977 |
0.977 |
0.953 |
| RFN |
64.2 |
0.342 |
0.309 |
0.302 |
0.433 |
0.477 |
| Rhino_CRE |
81.4 |
0.581 |
0.581 |
0.976 |
0.976 |
0.750 |
| S_box |
56.5 |
0.430 |
0.430 |
0.817 |
0.860 |
0.750 |
| s2m |
78.3 |
1.000 |
1.000 |
1.000 |
1.000 |
1.000 |
| SECIS |
36.5 |
0.000 |
0.000 |
0.000 |
0.000 |
-0.003 |
| Tymo_tRNA-like |
64.1 |
0.691 |
0.703 |
0.768 |
0.768 |
0.596 |
| |
Mean |
0.599 |
0.631 |
0.735 |
0.737 |
0.715 |
| |
Specificity |
0.947 |
0.947 |
0.947 |
0.947 |
0.999 |
| Sensitivity | 0.486 | 0.545 | 0.689 | 0.704 | 0.655 |
The MCC on non-structural alignments of the CMfinder dataset is compared between PhyloRNAalifold, RNAalifold and PETfold. The parameter β of PhyloRNAalifold is 10. Best performance on the same family is set to bold.