Table 3 Performance of QSAR based predictive models developed on each of the HIV protein inhibitor (IC50) datasets during tenfold training/testing and on independent validation
From: HIVprotI: an integrated web based platform for prediction and design of HIV proteins inhibitors
Serial number | HIV protein | Inhibitor compounds | Number of selected descriptors | Pearson’s correlation coefficient (PCC) | p value | |||
|---|---|---|---|---|---|---|---|---|
Total | Training/testing | Independent validation | Training/testing (10×) | Independent validation | ||||
1 | Protease | 1895 | 1706 | 189 | 45 | 0.78 | 0.73 | 1.00e−9 |
2 | Reverse transcriptase | 2126 | 1914 | 212 | 61 | 0.76 | 0.72 | 1.00e−7 |
3 | Integrase | 1240 | 1116 | 124 | 55 | 0.74 | 0.70 | 1.00e−6 |