Table 3.
Comparison of the best QMEAN versions with other methods participating in CASP7.
| Scoring function | Pearson | Spearman | sum(GDT) |
| QMEAN | 0.752 | 0.684 | 56.70 |
| Circle-QA | 0.718 | 0.643 | 56.03 |
| ProQ | 0.700 | 0.571 | 54.29 |
| ProQlocal | 0.698 | 0.563 | 54.17 |
| Bilab | 0.683 | 0.561 | 54.50 |
| ModFOLD | 0.661 | 0.580 | 54.19 |
| ABIpro | 0.653 | 0.605 | 56.40 |
| selfQMEAN | 0.830 | 0.749 | 56.60 |
| QMEANclust | 0.892 | 0.841 | 58.11 |
| Pcons | 0.801 | 0.714 | 54.36 |
| TASSER-QA | 0.828 | 0.785 | 57.23 |
| Zhang server | - | - | 57.35 |
| Random model selection | - | - | 46.19 |
| Best model per target | - | - | 62.00 |
Average correlation coefficient and total maximum GDT_TS score of the optimised QMEAN, QMEANclust and selfQMEAN versions and the top performing methods of CASP7. Only scoring functions with predictions for all 98 targets are shown.