Click on the links below to view different info pages.

If you have more questions, do not hesitate to ask


As an input, the VoroMQA server accepts one or more structures (models) in PDB format. Input structures can contain multiple chains, biological assemblies are also accepted. User can provide a sequence to filter and renumber the residues in the submitted PDB files. User can also enable the evaluation of inter-chain interface in addition to the whole structure assessment.


As an output, the server provides global scores, local (per-residue) scores, and additional local context information on secondary structure and solvent accessibility. Moreover, if the evaluation of inter-chain interactions was requested, the server provides interface quality scores, interface energy estimates, and local scores for residues involved in inter-chain interfaces.

All the local scores are presented in forms of interactive (clickable) plots and interactive color-coded scoring profiles and 3D structures. Various visualizations can be turned off and on, allowing a user to focus on some of the features without being distracted by the others, which is particularly useful when viewing results for multiple models on a single page. The VoroMQA server also provides an interactive plot for viewing and interpreting global scores of multiple models.

Context help

Almost every page contains a "Show help" button. It can be clicked to view the appropriate instructions and hints for the page.

Interpreting global VoroMQA scores

  • A vast majority of high quality experimentally determined structures have VoroMQA scores greater than 0.4.
  • A relatively very small fraction of the native structures have VoroMQA scores less than 0.3.
  • The plot below shows the distribution of the VoroMQA global scores of high-quality PDB structures:
  • For quickly interpreting a global VoroMQA score of a structural model, the following simple rule can be used:
    • If the score is greater than 0.4, then the model is likely good.
    • If the score is less than 0.3, then the model is likely bad.
    • If the score is between 0.3 and 0.4, then the model cannot be reliably classified as either good or bad based on just VoroMQA.