Skip to main content

Table 1 Performance of proteochemometric models for HIV-1 protease drug susceptibility predictions.

From: Proteochemometric modeling of HIV protease susceptibility

Model

Descriptor blocks

Goodness of fit (R2)

Predictive ability (Q2)

RMSEP*

Results of permutation test

     

R2 intercept

Q2 intercept

Model-1

P+I

0.75

0.72

0.44

0.02

-0.08

Model-2

P+I, P × I

0.86

0.82

0.35

0.14

-0.19

Model-3

P+I, P × I, P × P

0.91

0.87

0.30

0.21

-0.27

  1. * Root mean Squared Errors of Prediction