Upload 1–200 predicted, designed, or experimental models (PDB or CIF), up to 5,000 protein residues each. Each model is graded on its own (first model, selected chains), coordinates unchanged. Select a file to preview it.
Heavy atoms are graded against CogniChem-defined Ramachandran regions, canonical side-chain rotamers, van der Waals overlap, peptide ω, and ideal Cβ placement. Each ligand's bonds and angles are compared with an MMFF94-minimized copy of itself. Every metric is marked pass, warn, or fail against the thresholds below; values do not numerically match MolProbity.
Leave a field empty to use the default shown in it. A metric at or above its warn value is a warning and at or above its fail value a failure; Ramachandran favored grades low values (at or below).