Table 4 Summary of different quality evaluation parameters for the best model structure for each of the selected protein.
From: Benchmarking protein language models for protein crystallization
Quality parameters | Prot-142 | Prot-630 | Prot-851 | Prot-1120 | Prot-1302 |
|---|---|---|---|---|---|
Ramachandran distribution | Â | Â | Â | Â | Â |
Core region | 98.9% | 98.1% | 98.8% | 98.5% | 97.4% |
Allowed region | 1.1% | 1.9% | 1.2% | 1.1% | 2.6% |
Generously allowed region | 0.0% | 0.0% | 0.0% | 0.4% | 0.0% |
Disallowed region | 0.0% | 0.0% | 0.0% | 0.0% | 0.0% |
Bond lengths within limits | 96.6% | 97.3% | 96.6% | 97.9% | 96.5% |
Bond angles within limits | 93.8% | 94.3% | 92.6% | 94.0% | 92.9% |
Planner groups within limits | 98.8% | 100.0% | 98.1% | 100.0% | 99.0% |
Favored rotamers | 98.7% | 98.6% | 97.8% | 97.1% | 99.5% |
Errat score | 97.3% | 99.0% | 98.1% | 98.0% | 96.4% |
MolProbity score | 1.72 | 1.21 | 1.57 | 1.31 | 1.40 |
Predicted GDT-TS Score | 0.75 | 0.84 | 0.63 | 0.72 | 0.88 |
Predicted TM-score | 0.83 | 0.83 | 0.65 | 0.80 | 0.82 |
Global quality score | 0.44 | 0.41 | 0.45 | 0.44 | 0.35 |
Average quality score | 0.67 | 0.69 | 0.58 | 0.65 | 0.69 |
