# How to Use MolProbity for Protein Structure Validation: A Step-by-Step Tutorial

## Direct Answer and Scope

MolProbity is a web-based structure validation service that evaluates the geometric and stereochemical quality of macromolecular models, primarily protein structures determined by X-ray crystallography, cryo-electron microscopy, and nuclear magnetic resonance spectroscopy. This tutorial provides a practical workflow for submitting a PDB-format coordinate file to the MolProbity server, interpreting the validation report, and using the results to guide model improvement before deposition or further analysis. The intended readers are biology students, structural biology researchers, laboratory professionals, and life-science practitioners who need to validate their own structures or assess the quality of models downloaded from public repositories. The scope covers the complete pipeline from file preparation through output interpretation, with emphasis on the metrics that matter for publication and downstream use in molecular docking and structure-based drug design.

## Context and Importance of Structure Validation

### Why Validation Matters in Structural Biology

Protein structure determination produces atomic coordinates that are models derived from experimental data, not direct observations of every atom position. The quality of these models varies substantially depending on the resolution of the experimental data, the refinement strategy used, and the skill of the model builder. A structure with poor geometry can mislead downstream analyses including molecular docking, mutagenesis design, and mechanistic interpretation. Validation tools assess whether a model conforms to known chemical and physical constraints, such as bond lengths, bond angles, planar groups, and the avoidance of atomic clashes.

Public structure repositories such as the Protein Data Bank maintain quality standards, and the National Center for Biotechnology Information provides access to structure data and related bioinformatics resources that researchers use to retrieve and compare validated models. The NCBI data resources include search systems and analysis services that support structure-based research, and understanding the quality metrics of deposited structures is essential when using these resources for comparative studies.

### The Role of MolProbity in the Validation Ecosystem

MolProbity provides a unified scoring system that combines multiple geometric criteria into a single overall score. The server evaluates Ramachandran plot outliers, rotamer outliers, C-beta deviations, and atomic clashes, then reports a MolProbity score that combines these metrics into a percentile-based ranking relative to structures of similar resolution. This score allows researchers to compare their model against the distribution of deposited structures and to track improvements during refinement.

The validation output is used by journals and deposition databases as a quality gate. Many journals require validation reports to accompany structure submissions, and the wwPDB validation pipeline incorporates MolProbity-style metrics. Understanding how to generate and interpret these reports is therefore a practical skill for any researcher who plans to deposit a structure or who needs to evaluate the reliability of a model before using it in computational studies.

## Core Principles of MolProbity Scoring

### Geometric Criteria Evaluated by MolProbity

MolProbity evaluates several distinct aspects of model geometry. The Ramachandran plot analysis assesses the backbone dihedral angles of each residue against allowed and favored regions derived from high-resolution structures. Residues in disallowed regions are flagged as outliers because their backbone conformation is energetically unfavorable. The rotamer analysis evaluates side-chain dihedral angles against a library of preferred conformations, flagging residues whose side chains adopt rare or impossible arrangements.

The clash score quantifies the number of steric overlaps between atoms that are not bonded to each other. Two atoms are considered clashing when their van der Waals spheres overlap by more than a threshold amount. The clashscore is reported as the number of clashes per thousand atoms, normalized so that scores from different structures can be compared. The C-beta deviation metric identifies residues where the C-beta atom position deviates from the ideal position calculated from the backbone geometry, which often indicates a backbone trace error.

### The MolProbity Score and Percentile Ranking

The MolProbity score combines the clashscore, percentage of Ramachandran outliers, and percentage of rotamer outliers into a single weighted score. This score is then compared against a reference distribution of structures determined at similar resolution, producing a percentile ranking. A structure with a MolProbity score in the 90th percentile is better than 90 percent of comparable structures in the reference set.

The score is resolution-dependent because lower-resolution structures inherently have more geometric uncertainty. A structure determined at 3.5 angstrom resolution will typically have a higher MolProbity score than a structure determined at 1.5 angstrom resolution, even when both are well refined. The percentile ranking accounts for this by comparing each structure only against others in the same resolution bin.

### Relationship to Other Validation Metrics

MolProbity complements other validation approaches that assess model quality from different perspectives. The fit of the model to the experimental density map is evaluated separately by tools that calculate correlation coefficients or real-space R-factors. A model can have excellent geometry but poor density fit, or excellent density fit but poor geometry. Both aspects must be considered when assessing overall model quality.

For cryo-EM structures, the modeling process often involves building atomic coordinates into density maps, and recent work has shown that ensemble approaches using generative AI combined with simulation-based refinement can improve fitting accuracy for membrane proteins that undergo conformational transitions between functional states. The resulting models still require geometric validation, and MolProbity provides the stereochemical quality assessment that complements the map-fit evaluation.

## Preparing Your Structure File for MolProbity

### Required File Format and Content

MolProbity accepts coordinate files in PDB format, the standard text-based format for macromolecular structures. The file must contain atomic coordinate records with proper atom names, residue names, chain identifiers, and occupancy and B-factor fields. The server also accepts mmCIF format files, which are increasingly the standard for structure deposition.

Before submitting a file, verify that the structure is complete and that all atoms are present. Missing atoms, particularly in side chains, will be flagged during validation and may produce misleading results. If the structure contains ligands, water molecules, or other heteroatoms, these should be included in the file because they contribute to the clash analysis. However, the validation metrics for protein geometry are calculated on the protein atoms, and the presence of poorly fitted ligands can inflate the clashscore.

### Checking File Integrity Before Submission

Several preliminary checks can prevent common submission errors. Verify that the file contains no formatting errors, such as truncated lines or incorrect atom name alignment. Confirm that all residue names are standard three-letter codes and that chain identifiers are consistent throughout the file. Check that alternate conformations are properly labeled with alternate location indicators and that the occupancy values for alternate conformers sum to one.

The Carpentries lessons provide foundational training in data handling and file management that is useful for researchers who need to manipulate structure files programmatically. Basic command-line skills for inspecting and reformatting files can save substantial time when preparing multiple structures for validation.

### Handling Structures with Multiple Chains and Ligands

For multi-chain structures, MolProbity analyzes each chain separately and also reports whole-structure metrics. The per-chain breakdown is useful for identifying whether geometric problems are concentrated in a particular region of the complex. Ligands and nucleic acids are included in the clash analysis but are not scored by the protein-specific metrics.

When a structure contains covalent ligands or modified residues, the linkage records in the PDB file must be correct for the validation to be meaningful. Incorrect linkage records can produce spurious clashes or missed clashes at the modification site. Review the connectivity records in the file before submission and correct any errors.

## Step-by-Step Workflow for Running MolProbity

### Step 1: Access the MolProbity Web Server

Navigate to the MolProbity web server using a standard web browser. The server is maintained by the Richardson laboratory at Duke University and is freely available for academic use. The interface accepts file uploads and provides options for specifying the type of validation to perform.

The server may require registration or may allow anonymous use depending on the current implementation. If registration is required, create an account and log in before submitting structures. The server processes jobs asynchronously, and results are typically available within minutes for structures of moderate size.

### Step 2: Upload the Coordinate File

Use the file upload interface to submit your PDB or mmCIF file. The server accepts files up to a size limit, which is sufficient for most protein structures and complexes. After uploading, the server reads the file and displays basic information about the structure, including the number of chains, residues, and atoms.

Verify that the server has correctly parsed the file by checking the displayed summary against your expectations. If the server reports an error, review the file format and correct any issues before resubmitting. Common errors include incorrect atom naming, missing chain identifiers, and malformed coordinate records.

### Step 3: Select Validation Options

MolProbity offers several validation options that can be selected before running the analysis. The default options include clash analysis, Ramachandran plot analysis, rotamer analysis, and C-beta deviation analysis. Additional options include the analysis of nucleic acid geometry for structures containing DNA or RNA, and the analysis of ligand geometry for structures containing small molecules.

For most validation purposes, the default options are appropriate. If you are validating a structure for deposition, use the full validation suite to generate a complete report. If you are performing a quick check during refinement, the default options provide the essential metrics without the additional computation time.

### Step 4: Run the Validation and Retrieve Results

After selecting the validation options, submit the job and wait for the results. The server displays a progress indicator and redirects to the results page when the analysis is complete. The results page contains a summary of all validation metrics, including the MolProbity score, clashscore, Ramachambandran statistics, and rotamer statistics.

Download the full validation report for your records. The report includes detailed lists of outliers that can be used to guide model rebuilding. The server also provides the option to download the multi-criteria validation file in a machine-readable format for integration into automated pipelines.

### Step 5: Interpret the Summary Report

The summary report presents the key metrics in a table format with color coding that indicates the quality of each metric. Green indicates good quality, yellow indicates caution, and red indicates poor quality. The MolProbity score is presented with its percentile ranking, allowing you to assess how your structure compares to the reference distribution.

The clashscore is reported as the number of clashes per thousand atoms. Lower values are better, and a clashscore below 5 is generally considered good for structures determined at moderate resolution. The Ramachandran analysis reports the percentage of residues in favored and allowed regions, with outliers listed separately. The rotamer analysis reports the percentage of residues with preferred rotamers and lists the outliers.

## At a Glance: MolProbity Metrics and Interpretation

| Metric | What It Measures | Good Range | Action If Poor |
|--------|-----------------|------------|----------------|
| MolProbity Score | Combined geometric quality percentile | 90th percentile or higher | Review individual component metrics and rebuild problem regions |
| Clashscore | Steric overlaps per thousand atoms | Below 5 for moderate resolution | Identify clash clusters and adjust side chains or backbone |
| Ramachandran Outliers | Residues in disallowed backbone conformations | Below 0.5 percent | Inspect outlier residues and rebuild backbone if density supports |
| Rotamer Outliers | Side chains in rare conformations | Below 1 percent | Replace outlier rotamers with preferred conformations |
| C-beta Deviations | Backbone trace errors indicated by C-beta position | Fewer than 5 deviations | Check local density fit and rebuild the affected region |

## Interpreting the Detailed Validation Report

### Ramachandran Plot Analysis

The Ramachandran plot is a two-dimensional histogram of the backbone dihedral angles phi and psi for each residue. MolProbity generates the plot and marks each residue as favored, allowed, or outlier based on its position relative to reference distributions. The favored and allowed regions are defined using a large dataset of high-resolution structures, and the boundaries are chosen to include the vast majority of residues in well-determined structures.

Residues in the outlier region are flagged because their backbone conformation is rarely observed in high-quality structures. However, not all outliers are errors. Some residues, particularly glycine and proline, have different Ramachandran preferences than other amino acids. Glycine is more flexible because its side chain is a hydrogen atom, and proline is constrained by its cyclic side chain. MolProbity accounts for these differences by using residue-specific reference distributions.

When a Ramachandran outlier is identified, examine the local electron density or cryo-EM map to determine whether the backbone conformation is supported by the experimental data. If the density clearly supports the outlier conformation, the residue may be genuinely strained, and the outlier flag can be justified in the deposition. If the density does not support the conformation, rebuild the backbone to a more favorable geometry.

### Rotamer Analysis

Rotamers are the preferred conformations of amino acid side chains, defined by the dihedral angles chi1 and chi2. MolProbity compares each side chain against a rotamer library derived from high-resolution structures and flags residues whose side-chain conformations are rare or unobserved.

Rotamer outliers often indicate errors in side-chain placement that can be corrected by rebuilding the side chain into the density. In some cases, the rotamer outlier is caused by a neighboring residue that is incorrectly placed, and correcting the neighbor resolves the problem. In other cases, the side chain is genuinely strained by interactions with the surrounding environment, such as in enzyme active sites where the side chain is held in an unusual conformation for catalysis.

The rotamer analysis report lists each outlier with its residue number, chain, and the observed and preferred rotamers. Use this information to prioritize rebuilding efforts. Residues with severe rotamer outliers that are not supported by density should be corrected before deposition.

### Clash Analysis

The clash analysis identifies pairs of atoms that overlap sterically. MolProbity calculates the van der Waals radii of all atoms and flags pairs whose interatomic distance is less than the sum of their radii minus a tolerance. The clashscore is the number of such clashes per thousand atoms.

Clashes can occur between atoms in the same residue, between adjacent residues, between different chains, or between protein and ligand atoms. The clash report lists each clash with the two atoms involved and the distance between them. Clusters of clashes often indicate a region where the model is locally incorrect, such as a misregistered loop or a side chain placed in the wrong rotamer.

When correcting clashes, prioritize the most severe overlaps first. Adjust side-chain rotamers to relieve clashes, and if the clash persists, examine the backbone conformation and the density fit. In some cases, the clash is caused by a ligand that is poorly fitted, and rebuilding the ligand resolves the problem.

### C-beta Deviation Analysis

The C-beta deviation metric identifies residues where the C-beta atom position deviates from the position calculated from the backbone geometry. A large deviation indicates that the backbone trace is locally incorrect, often because the residue has been misregistered in the sequence or because the backbone has been traced through the density incorrectly.

C-beta deviations are particularly useful for identifying register errors in cryo-EM models, where the density may be ambiguous and the sequence can be shifted by one or more residues. When a C-beta deviation is identified, examine the density in the region and verify that the residue assignment is correct. If the residue is misassigned, rebuild the region with the correct sequence register.

## Using MolProbity During Model Refinement

### Iterative Validation and Rebuilding

MolProbity is most useful when used iteratively during refinement instead of as a final check before deposition. After each round of rebuilding, run MolProbity to assess whether the geometric quality has improved. Track the MolProbity score, clashscore, and outlier counts across refinement cycles to document progress.

The iterative workflow involves identifying problem regions from the validation report, rebuilding those regions into the density, and re-running validation to confirm improvement. This cycle is repeated until the validation metrics reach acceptable levels or until further improvement is not possible given the resolution and quality of the experimental data.

### Prioritizing Corrections

Not all validation outliers require correction. The priority for correction depends on the severity of the problem and the strength of the experimental evidence. Severe clashes and Ramachandran outliers that are not supported by density should be corrected first. Minor rotamer outliers that are supported by density may be acceptable if the density clearly indicates the observed conformation.

The validation report provides a ranked list of problems, with the most severe issues listed first. Use this ranking to guide the rebuilding effort. Correcting the most severe problems first often resolves multiple minor issues because the corrections improve the local environment for neighboring residues.

### Tracking Validation Metrics Across Cycles

Maintain a record of validation metrics for each refinement cycle. The record should include the MolProbity score, clashscore, Ramachandran outlier percentage, rotamer outlier percentage, and the number of C-beta deviations. This record documents the improvement during refinement and provides evidence of the model quality for the deposition.

The record also helps identify when refinement has converged. When the validation metrics stop improving despite additional rebuilding, the model has reached the quality limit imposed by the experimental data. Further refinement is unlikely to produce substantial improvement and may risk overfitting the model to noise.

## Common Failure Patterns and Troubleshooting

### Failure Pattern 1: High Clashscore in a Specific Region

A high clashscore concentrated in one region of the structure often indicates a local modeling error. The region may contain a misregistered loop, an incorrectly placed side chain, or a ligand that is poorly fitted. Examine the density in the region and compare the model to the density to identify the source of the clashes.

Correct the modeling error and re-run validation. If the clashes persist after correction, the problem may be in the refinement strategy instead of the model itself. Consider using a different refinement protocol or adding restraints to the problem region.

### Failure Pattern 2: Many Ramachandran Outliers

A large number of Ramachandran outliers suggests a systematic problem with the backbone geometry. This can occur when the model has been refined with weak geometric restraints, when the density is poor in the affected regions, or when the sequence has been misregistered.

Check the resolution of the structure and the quality of the density in the outlier regions. If the density is poor, the outliers may reflect genuine uncertainty in the backbone position. If the density is good, rebuild the backbone to fit the density with favorable geometry.

### Failure Pattern 3: Rotamer Outliers in the Core

Rotamer outliers in the protein core are often caused by incorrect side-chain packing. The side chain may be placed in a conformation that avoids clashes with neighboring atoms but is not a preferred rotamer. Alternatively, the side chain may be in a preferred rotamer but the neighboring residues are incorrectly placed, creating a false outlier.

Examine the packing environment around the outlier residue. If the neighboring side chains are correctly placed, rebuild the outlier side chain to a preferred rotamer that fits the density. If the neighbors are incorrect, correct the neighbors first and then re-evaluate the outlier.

### Failure Pattern 4: Poor MolProbity Score Despite Good Individual Metrics

A poor MolProbity score can occur when the individual metrics are acceptable but the structure is at a resolution where the reference distribution is demanding. The percentile ranking compares the structure against others at similar resolution, and a structure at the low end of the resolution range may rank poorly even with reasonable geometry.

Check the resolution of the structure and the distribution of metrics in the reference set. If the structure is at a resolution where few comparable structures exist, the percentile ranking may be unreliable. Focus on the absolute values of the metrics instead of the percentile in this case.

## Limitations and Interpretation Boundaries

### Resolution Dependence of Validation Metrics

MolProbity metrics are resolution-dependent, and the interpretation of a given score depends on the resolution of the structure. A clashscore of 10 may be acceptable for a structure at 3.5 angstrom resolution but poor for a structure at 2.0 angstrom resolution. The percentile ranking accounts for this by comparing against structures in the same resolution bin, but the absolute values should still be interpreted with the resolution in mind.

When comparing structures at different resolutions, use the percentile rankings instead of the raw scores. A structure at 3.0 angstrom resolution with a MolProbity score in the 80th percentile is comparable in quality to a structure at 2.0 angstrom resolution with a score in the 80th percentile, even though the raw scores differ.

### What MolProbity Does Not Assess

MolProbity assesses geometric quality but does not assess the fit of the model to the experimental data. A model can have excellent geometry and be completely wrong if it has been refined against incorrect data or if the density has been misinterpreted. The fit of the model to the density must be evaluated separately using tools that calculate map-model correlation or real-space R-factors.

MolProbity also does not assess the biological relevance of the model. A model can have excellent geometry and be biologically meaningless if the crystallization conditions or the cryo-EM preparation have produced an artifact. The biological interpretation of the structure requires additional analysis beyond geometric validation.

### The Role of Human Judgment

Validation metrics provide quantitative assessments of model quality, but they do not replace human judgment. A residue flagged as a Ramachandran outlier may be genuinely strained and biologically important, such as a catalytic residue held in an unusual conformation. The decision to correct or retain an outlier should be based on the experimental evidence and the biological context, not solely on the validation score.

When a validation outlier is retained, document the justification in the deposition. The validation report can be annotated to explain why specific outliers are retained, and this annotation is reviewed by the deposition database and by journal reviewers.

## Integration with Other Bioinformatics Resources

### Using NCBI Resources for Structure Comparison

The National Center for Biotechnology Information provides access to structure databases and analysis tools that complement MolProbity validation. Researchers can retrieve structures from the Protein Data Bank through NCBI's search systems and compare their validation metrics against the distribution of deposited structures. The NCBI resources also provide access to sequence data and annotations that support the interpretation of structure quality in a biological context.

When validating a structure, compare the MolProbity metrics against those of related structures in the database. If the new structure has substantially worse metrics than related structures at similar resolution, the model may require additional refinement. If the metrics are comparable, the model is likely of acceptable quality.

### Training Resources for Structure Validation Skills

The EMBL-EBI Training program provides learning pathways for bioinformatics analysis, including structure validation and quality assessment. These training resources are useful for researchers who are new to structure validation and need to understand the principles behind the metrics. The training materials cover the practical aspects of using validation tools and interpreting their output.

The Galaxy Training Network offers accessible workflow training that includes structure analysis and validation. These tutorials provide hands-on experience with validation tools and demonstrate how to integrate validation into reproducible analysis workflows. The nf-core documentation describes community standards for reproducible pipelines, which is relevant for researchers who want to incorporate MolProbity into automated analysis pipelines.

### Reproducible Validation Workflows

For research projects that involve validating many structures, consider building a reproducible validation workflow. The Bioconductor project provides packages for structural biology analysis that can be integrated with MolProbity output. These packages support the statistical analysis of validation metrics across structure sets and the generation of publication-quality figures.

A reproducible workflow should record the version of MolProbity used, the input file, the validation options, and the output metrics. This record allows the validation to be repeated and verified by other researchers. The nf-core documentation provides guidance on building reproducible pipelines that meet community standards.

## Records and Measurements for Validation

### Maintaining a Validation Log

Keep a validation log for each structure that records the input file, the MolProbity version, the validation options, the date of the analysis, and the complete output metrics. The log should also record any corrections made to the model and the validation metrics after each correction. This log documents the refinement history and provides evidence of the model quality for the deposition.

The validation log is particularly important for structures that are part of a larger project, such as a series of mutants or a family of related proteins. The log allows the validation metrics to be compared across the series and identifies structures that require additional refinement.

### Recording Metrics for Deposition

When depositing a structure, record the MolProbity metrics in the deposition. The wwPDB validation pipeline generates a validation report that includes MolProbity-style metrics, and this report is part of the deposition. The metrics should be reported accurately and any outliers should be explained in the deposition comments.

The deposition database uses the validation metrics to assess the quality of the structure and to provide guidance to the depositor. Structures with poor validation metrics may be returned for additional refinement before deposition is accepted.

### Using Metrics for Publication

Journals increasingly require validation metrics to be reported in structure papers. The MolProbity score, clashscore, and Ramachandran statistics are commonly reported in the methods section and in the structure quality tables. Report the metrics accurately and describe the validation protocol used.

The validation metrics provide reviewers with an objective assessment of the model quality. A structure with good validation metrics is more likely to be accepted for publication than a structure with poor metrics, regardless of the biological interest of the findings.

## Professional Escalation Criteria

### When to Seek Additional Expertise

If the validation metrics do not improve despite repeated rebuilding, the problem may require expertise beyond the standard refinement workflow. Consider consulting a structural biology colleague who has experience with the specific type of structure or the specific validation problem. A crystallographer or cryo-EM specialist can provide guidance on refinement strategies and on the interpretation of validation outliers.

If the structure is part of a collaborative project, discuss the validation results with the collaboration team. The team may have insights into the biology of the system that explain the validation outliers, or they may have access to additional experimental data that can guide the refinement.

### When to Consider Data Collection or Processing Issues

Persistent validation problems can indicate issues with the experimental data instead of the model. If the density is poor in the regions where validation outliers occur, the data collection or processing may need to be revisited. For crystallographic data, this may involve collecting higher-resolution data or improving the scaling and merging. For cryo-EM data, this may involve collecting more particles or improving the classification and refinement.

If the data quality is the limiting factor, the validation metrics will not improve beyond a certain point regardless of the refinement effort. In this case, the structure should be deposited with the achievable quality and the limitations should be documented.

### When to Reconsider the Model Building Strategy

If the validation metrics are consistently poor across multiple refinement cycles, the model building strategy may be flawed. Consider using a different model building approach, such as automated building tools or a different starting model. For cryo-EM structures, recent methods that generate ensembles of models using generative AI and refine them with density-guided simulations can improve the fitting accuracy for proteins that undergo conformational transitions.

The choice of starting model can have a substantial impact on the final validation metrics. A starting model with poor geometry will require more refinement to achieve acceptable metrics than a starting model with good geometry. Consider using a well-refined homologous structure as the starting model when available.

## Safety and Ethical Context

### Responsible Use of Validation Tools

Validation tools are designed to assess model quality, not to guarantee correctness. A structure with excellent validation metrics can still contain errors, and a structure with poor metrics can still be biologically informative. Use validation metrics as one component of a comprehensive quality assessment that includes the fit to the experimental data and the biological plausibility of the model.

When reporting validation metrics, be transparent about the limitations of the assessment. Report the resolution of the structure, the validation protocol used, and any known issues with the model. This transparency allows other researchers to assess the reliability of the structure for their own purposes.

### Data Integrity in Structure Deposition

Depositing a structure with known errors is a breach of scientific integrity. Before deposition, ensure that the model has been refined to the best achievable quality and that all validation outliers have been examined and either corrected or justified. The deposition should include an accurate description of the model quality and any limitations.

The validation report generated by the deposition database provides an independent assessment of the model quality. Review this report carefully and address any issues raised before the deposition is finalized.

### Educational Use of Validation Tools

MolProbity and other validation tools are valuable educational resources for teaching structural biology. Students can learn about protein geometry, stereochemistry, and the relationship between model quality and experimental data by validating structures and interpreting the output. The training resources from the EMBL-EBI Training program and the Galaxy Training Network provide structured learning pathways for this purpose.

When using validation tools in teaching, emphasize that the metrics are indicators of quality, not absolute measures of correctness. Students should learn to interpret the metrics in the context of the experimental data and the biological question being addressed.

## Frequently Asked Questions

### What file formats does MolProbity accept for structure validation?

MolProbity accepts coordinate files in PDB format and mmCIF format. The PDB format is the traditional text-based format with fixed column positions for atom records, while the mmCIF format is a more flexible key-value format that is increasingly the standard for structure deposition. Both formats must contain complete atomic coordinate records with proper atom naming, residue naming, and chain identifiers. Before submission, verify that the file is correctly formatted and that all required fields are present.

### How long does a MolProbity validation job take to complete?

The processing time depends on the size of the structure and the current load on the server. Small structures with a few hundred residues typically complete within a few minutes. Large complexes with multiple chains and many ligands may take longer. The server displays a progress indicator and provides a link to the results when the job is complete. If the job takes an unusually long time, check the server status and consider resubmitting the file.

### What is a good MolProbity score for a protein structure?

A good MolProbity score depends on the resolution of the structure. The score is reported as a percentile ranking relative to structures at similar resolution, and a score in the 90th percentile or higher is generally considered good. For a structure at 2.0 angstrom resolution, a MolProbity score below 1.5 is typically achievable with careful refinement. For a structure at 3.0 angstrom resolution, a score below 2.5 is reasonable. The percentile ranking is more informative than the raw score when comparing structures at different resolutions.

### How do I correct Ramachandran outliers in my structure?

To correct a Ramachandran outlier, first examine the electron density or cryo-EM map in the region of the outlier residue. If the density supports an alternative backbone conformation with favorable geometry, rebuild the backbone to match the density. If the density does not support the outlier conformation, the residue may be misassigned or the backbone may be traced incorrectly. In this case, rebuild the region with the correct sequence register and backbone trace. After rebuilding, re-run MolProbity to confirm that the outlier has been resolved.

### Can MolProbity validate cryo-EM structures?

Yes, MolProbity can validate cryo-EM structures. The geometric criteria assessed by MolProbity are independent of the experimental method used to determine the structure. Cryo-EM structures are validated using the same Ramachandran, rotamer, clash, and C-beta deviation metrics as crystallographic structures. However, cryo-EM structures may have different resolution distributions than crystallographic structures, and the percentile ranking accounts for this by comparing against structures at similar resolution.

### What should I do if my structure has a high clashscore?

A high clashscore indicates that many atoms are overlapping sterically. Examine the clash report to identify the regions where clashes are concentrated. Correct the most severe clashes first by adjusting side-chain rotamers or rebuilding the backbone. If the clashes are caused by a ligand, rebuild the ligand into the density. After correcting the clashes, re-run MolProbity to confirm that the clashscore has improved. If the clashscore remains high, the problem may be in the refinement strategy or the data quality.

### How does MolProbity handle ligands and nucleic acids?

MolProbity includes ligands and nucleic acids in the clash analysis but does not score them with the protein-specific metrics. The Ramachandran, rotamer, and C-beta deviation metrics apply only to protein residues. Nucleic acid geometry is assessed separately if the structure contains DNA or RNA. Ligand geometry can be assessed using additional tools that evaluate the fit of the ligand to the density and the geometric quality of the ligand conformation.

### Is MolProbity suitable for validating predicted structures from AlphaFold?

MolProbity can be used to validate predicted structures, but the interpretation of the metrics differs from experimental structures. Predicted structures are not refined against experimental data, so the geometric quality reflects the prediction method instead of the fit to data. The MolProbity metrics can identify geometric problems in predicted structures, but the percentile ranking is based on experimental structures and may not be directly applicable. For predicted structures, focus on the absolute values of the metrics and on the identification of problematic regions instead of the percentile ranking.

## Related Bioinformatics Guides

- [Gene Set Enrichment Analysis in R: A Practical Tutorial for Interpreting Omics Data](/knowledge/bioinformatics/gene-set-enrichment-analysis-in-r-a-practical-tutorial-for-interpreting-omics-data)
- [Protein Language Models in Bioinformatics: A Practical Guide to Selection and Application](/knowledge/bioinformatics/protein-language-models-in-bioinformatics-a-practical-guide-to-selection-and-application)
- [Digital Pathology Validation: A Practical Guide to CAP and RCPath Compliance](/knowledge/bioinformatics/digital-pathology-validation-a-practical-guide-to-cap-and-rcpath-compliance)
- [Structural Comparison and Alignment Algorithms for Protein 3D Structures](/knowledge/bioinformatics/structural-comparison-and-alignment-algorithms-for-protein-3d-structures)
- [How To Use Alphafold To Predict Structure: Structural Analysis and Computational Methodologies in Bioinformatics](/knowledge/bioinformatics/how-to-use-alphafold-to-predict-structure)

## References and Further Reading

- [NCBI Data Resources](https://www.ncbi.nlm.nih.gov/). National Center for Biotechnology Information.
- [EMBL-EBI Training](https://www.ebi.ac.uk/training). European Bioinformatics Institute.
- [Bioconductor](https://bioconductor.org/). Bioconductor Project.
- [Galaxy Training Network](https://training.galaxyproject.org/). Galaxy Project.
- [nf-core Documentation](https://nf-co.re/docs). nf-core.
- [The Carpentries Lessons](https://carpentries.org/lessons). The Carpentries.
- [Modeling cryo-EM structures in alternative states with AlphaFold2-based models and density-guided simulations.](https://doi.org/10.1038/s42004-025-01751-4). 2025.
- [Protocol for designing a peptide-based multi-epitope vaccine targeting monkeypox using reverse vaccine technology.](https://doi.org/10.1016/j.xpro.2025.103671). 2025.
- [Building molecular model series from heterogeneous CryoEM structures using Gaussian mixture models and deep neural networks.](https://doi.org/10.1038/s42003-025-08202-9). 2025.
- [A Tutorial on Movable Antennas for Wireless Networks](https://doi.org/10.1109/comst.2025.3546373). IEEE Communications Surveys and Tutorials, 2025.
- [Pinching-Antenna Systems (PASS): A Tutorial](https://doi.org/10.1109/TCOMM.2026.3658289). IEEE Transactions on Communications, 2025.

> This article is educational and does not replace validated analysis plans, institutional policy, clinical interpretation, or specialist review.