Table 3 |
|||||||||||
|
Comparison of the performances of the best sequence-based SVM method (SVM-WIN31) and structure-based SVM method (SVM-3D9) with the I-Mutant based predictors. |
|||||||||||
|
Method |
<Q> |
<C> |
Q[-] |
P[-] |
C[-] |
Q[N] |
P[N] |
C[N] |
Q[+] |
P[+] |
C[+] |
|
|
|||||||||||
|
I-Mutant Seq |
0.52 |
0.13 |
0.59 |
0.46 |
0.34 |
0.73 |
0.55 |
0.11 |
0.00 |
0.00 |
−0.04 |
|
SVM-WIN31 |
0.55 |
0.24 |
0.45 |
0.51 |
0.33 |
0.64 |
0.58 |
0.16 |
0.45 |
0.51 |
0.32 |
|
I-Mutant 3D |
0.54 |
0.22 |
0.70 |
0.43 |
0.35 |
0.72 |
0.61 |
0.25 |
0.00 |
0.33 |
0.01 |
|
SVM-3D9 |
0.58 |
0.33 |
0.55 |
0.51 |
0.37 |
0.62 |
0.67 |
0.29 |
0.55 |
0.51 |
0.37 |
|
|
|||||||||||
|
For notation see Table 1. I-Mutant SEQ, I-Mutant 3D, SVM-WIN31 and SVM-3D9 are tested on 982 mutations of NewDB database. |
|||||||||||
|
Capriotti et al. BMC Bioinformatics 2008 9(Suppl 2):S6 doi:10.1186/1471-2105-9-S2-S6 |
|||||||||||