Table 3.
The top I-TASSER model (largest cluster) for 3qf4A and 3ug9A with or without using MemBrain contact predictions
| PDBID | Cutoff | RMSD/ TMa | RMSD/ TMb | RMSD/ TMc | RMSD/ TMd |
|---|---|---|---|---|---|
| 3qf4A | 30e | 4.3/0.607 | 4.2/0.610 | 6.1/0.715 | 6.0/0.716 |
| 15e | 4.2/0.605 | 4.2/0.606 | 25.3/0.437 | 23.0/0.449 | |
| A15f | 6.7/0.523 | 4.6/0.600 | 26.0/0.422 | 25.4/0.417 | |
| 50g | 16.7/0.264 | 16.5/0.261 | 22.2/0.288 | 22.0/0.286 | |
| 3ug9A | 30e | 3.0/0.795 | 3.2/0.765 | 15.0/0.649 | 14.9/0.642 |
| 15e | 5.6/0.524 | 4.2/0.599 | 12.5/0.450 | 11.9/0.500 | |
| A15f | 10.9/0.369 | 5.3/0.600 | 19.9/0.323 | 16.0/0.500 | |
| 50g | 15.0/0.219 | 6.9/0.564 | 17.4/0.231 | 16.3/0.472 |
Note: The RMSD and TM-score to native for different degrees of template structure removal are given.
aRMSD (Å) and TM-score in the TMH region without using MemBrain predictions.
bRMSD (Å) and TM-score in the TMH region using MemBrain predictions.
cRMSD (Å) and TM-score of the whole chain without using MemBrain predictions.
dRMSD (Å) and TM-score of the whole chain using MemBrain predictions.
eTemplates whose sequence has >X% sequence identity to the protein were removed.
fAverage results of the top 10 models, where templates whose sequence has >15% sequence identity to the protein were removed.
gTemplates whose structure has TM-score >0.5 to the native structure were removed.