# How to Perform Conservation Analysis on Protein Structures: A Step-by-Step Guide Using ConSurf and PyMOL

Conservation analysis maps evolutionary sequence variability onto three-dimensional protein structures to identify residues that are likely critical for function, stability, or interaction. This guide provides a practical workflow for biology students, researchers, and laboratory professionals who need to combine ConSurf conservation scores with PyMOL structural visualization. The protocol covers data preparation, running conservation calculations, interpreting grade assignments, and producing publication-ready structural figures. The intended outcome is a reproducible method for locating functionally important residues in a protein of interest and for making defensible interpretations about their biological relevance.

## Scope and Reader Context

This workflow assumes you have a protein structure file, typically in PDB or mmCIF format, and a corresponding amino acid sequence. You will use the ConSurf web server to calculate per-residue conservation scores from a multiple sequence alignment of homologous proteins, then use PyMOL to color the structure by those scores. The method applies to experimentally determined structures from X-ray crystallography, NMR spectroscopy, or cryo-electron microscopy, as well as to high-confidence computational models. The protocol is designed for single-chain analysis but can be adapted for multimeric complexes with careful attention to chain identifiers.

Researchers use conservation mapping for several common purposes: identifying catalytic residues, locating binding interfaces, prioritizing mutations for functional studies, and interpreting the potential impact of disease-associated variants. The approach is complementary to other structural bioinformatics methods such as protein structure prediction, molecular docking interpretation, and biophysical feature analysis. Conservation data provide evolutionary evidence that can strengthen or challenge conclusions drawn from structural inspection alone.

## At a Glance

The table below summarizes the core workflow stages, the primary tools involved, and the expected outputs at each step.

| Workflow Stage | Primary Tool | Input Required | Output Produced |
| --- | --- | --- | --- |
| Structure preparation | NCBI Structure or RCSB | PDB ID or structure file | Cleaned coordinate file with chain and residue identifiers |
| Sequence retrieval | NCBI Protein | Protein name or accession | FASTA sequence matching the structure |
| Homolog search and alignment | ConSurf server | FASTA sequence | Multiple sequence alignment of homologous proteins |
| Conservation scoring | ConSurf server | Alignment and structure | Per-residue conservation grades from 1 to 9 |
| Structural visualization | PyMOL | Structure file and ConSurf output | Color-mapped structure with conserved regions highlighted |
| Interpretation and reporting | Manual analysis | Conservation grades and structural context | Identification of functionally important residues |

## Understanding Conservation Analysis

### Evolutionary Basis of Conservation Scoring

Conservation analysis rests on a simple evolutionary principle: amino acid positions that tolerate change across evolutionary time are usually not critical for function, while positions that remain identical or similar across diverse species are likely under purifying selection. A residue that appears in the same position across many distantly related homologs has probably been preserved because mutations at that position impair protein function. Conversely, positions that show many different amino acids across homologs are probably in regions where sequence variation is tolerated.

The statistical power of conservation analysis depends on the depth and diversity of the multiple sequence alignment. An alignment with sequences from many species spanning a broad evolutionary range provides stronger evidence than an alignment with only closely related sequences. The ConSurf server uses phylogenetic information to account for the evolutionary relationships among the aligned sequences, which prevents overrepresentation of closely related species from dominating the conservation signal.

### Relationship to Protein Structure and Function

Conserved residues tend to cluster in functionally important regions of proteins. Catalytic sites, ligand binding pockets, protein-protein interaction interfaces, and structural cores often show high conservation because mutations in these regions disrupt essential activities. Surface residues that face solvent and have no functional role typically show lower conservation because they can accommodate many different amino acids without consequences.

Structural context matters for interpretation. A conserved residue buried in the hydrophobic core may be important for folding stability, while a conserved residue on the surface may participate in binding or signaling. The same conservation grade can have different functional meanings depending on the local structural environment. Therefore, conservation scores should always be interpreted alongside structural features such as solvent accessibility, secondary structure, and proximity to known functional sites.

### Limitations of Conservation Data

Conservation analysis identifies positions under evolutionary constraint but does not directly reveal what function is being preserved. A conserved residue could be important for catalysis, binding, folding, allosteric regulation, or any other property that affects organism fitness. Additional experimental or computational evidence is needed to assign specific functional roles.

Conservation scores can also be misleading in certain situations. Proteins with very few homologs produce unreliable alignments and therefore unreliable scores. Proteins that have undergone functional divergence may show conservation patterns that reflect one function but not another. Convergent evolution can produce similar amino acids at positions that are not actually homologous. These limitations mean that conservation analysis should be treated as a hypothesis-generating tool instead of a definitive functional assay.

## Preparing Input Data

### Obtaining a Protein Structure

The first step is to obtain a three-dimensional structure for the protein of interest. The Protein Data Bank, accessible through the NCBI Structure database, is the primary repository for experimentally determined macromolecular structures. Search for your protein by name, organism, or PDB identifier. When multiple structures exist for the same protein, choose the one that best matches your research question. Consider resolution for crystallographic structures, the presence of ligands or binding partners, the biological assembly versus the asymmetric unit, and the completeness of the modeled regions.

For proteins without experimental structures, computational models can be used. Models generated by state-of-the-art prediction methods often have sufficient accuracy for conservation mapping, particularly in well-folded regions. However, be cautious about interpreting conservation in regions of low model confidence. The quality of the structure directly affects the reliability of the structural interpretation of conservation scores.

### Retrieving the Amino Acid Sequence

The sequence used for the conservation analysis must match the sequence of the structure. Retrieve the sequence from the NCBI Protein database using the protein name or accession number. Verify that the sequence corresponds to the same isoform and species as the structure. Differences between the sequence used for alignment and the sequence in the structure can cause mismatches in residue numbering and lead to incorrect conservation assignments.

When the structure contains mutations, tags, or other modifications not present in the wild-type sequence, decide whether to use the wild-type sequence or the construct sequence for the conservation analysis. For most applications, the wild-type sequence is preferred because it represents the naturally occurring protein. However, if the structure contains a mutation that is the focus of your study, you may want to analyze both sequences to understand how the mutation affects the conservation context.

### Formatting the Sequence for Submission

ConSurf accepts a single protein sequence in FASTA format. The FASTA header should contain a descriptive identifier, and the sequence should contain only standard amino acid one-letter codes. Remove any nonstandard characters, gaps, or numbering from the sequence. The server also accepts a multiple sequence alignment as input if you prefer to generate your own alignment using external tools.

Before submission, check the sequence length and composition. Very short sequences may not produce reliable conservation scores because too few informative positions exist. Very long sequences may require extended processing time. Sequences with unusual amino acid composition or low complexity regions may cause alignment difficulties.

## Running ConSurf Conservation Analysis

### Submitting a ConSurf Job

Navigate to the ConSurf web server and select the option to submit a new job. Paste your FASTA sequence into the submission form or upload a sequence file. Provide a valid email address to receive notification when the job is complete. The server assigns a job identifier that you can use to retrieve results later.

ConSurf offers several analysis options that affect the quality and interpretation of the results. The choice of database for homolog searching influences the number and diversity of sequences retrieved. The default settings are appropriate for most applications, but you may need to adjust parameters for proteins with unusual evolutionary histories or for analyses focused on specific taxonomic groups.

### Selecting Homolog Search Parameters

The homolog search identifies sequences from the protein database that are evolutionarily related to your query sequence. The sensitivity of the search depends on the number of iterations and the expectation value threshold. More sensitive searches retrieve more distant homologs, which can improve the evolutionary signal but also increase the risk of including sequences that are not true homologs.

The number of sequences included in the alignment affects the reliability of the conservation scores. Too few sequences provide insufficient statistical power. Too many sequences can introduce redundancy from closely related species. ConSurf provides options to control the maximum number of sequences and to filter sequences by sequence identity. A common approach is to include sequences with a range of identities that span the evolutionary diversity of the protein family.

### Choosing the Alignment Method

The multiple sequence alignment is the foundation of the conservation calculation. ConSurf offers several alignment programs, each with different strengths and weaknesses. The choice of alignment method can affect the final conservation scores, particularly in regions of the protein with insertions, deletions, or low sequence similarity.

For most proteins, the default alignment method produces acceptable results. However, for proteins with unusual features such as long disordered regions, repeated domains, or transmembrane segments, you may need to experiment with different alignment methods to obtain a biologically sensible alignment. Inspect the alignment visually before proceeding to the conservation calculation.

### Running the Conservation Calculation

After the alignment is complete, ConSurf calculates conservation scores for each position using a Bayesian or maximum likelihood approach that accounts for the phylogenetic relationships among the sequences. The method estimates the rate of evolution at each position and converts this rate into a conservation score.

The output includes a conservation grade from 1 to 9 for each residue, where 9 indicates the most conserved positions and 1 indicates the most variable positions. The grades are normalized so that they are comparable across different proteins. The server also provides a color scheme for visualizing the grades, with dark purple or burgundy representing highly conserved positions and turquoise or green representing variable positions.

### Retrieving and Interpreting Results

When the job is complete, you will receive an email with a link to the results page. The results page displays the conservation scores in several formats: a color-coded sequence alignment, a table of per-residue scores, and a structural visualization if you provided a structure file. Download the score table and the structural output files for use in PyMOL.

The results page also provides information about the alignment quality, including the number of sequences used and the distribution of sequence identities. Review this information to assess the reliability of the conservation scores. Alignments with very few sequences or with sequences that are all very similar to the query may produce less informative conservation scores.

## Preparing ConSurf Output for PyMOL

### Understanding the Output File Formats

ConSurf provides several output files that can be used for structural visualization. The most useful for PyMOL is the PDB file with conservation scores stored in the B-factor column. This file has the same coordinates as the original structure but with the B-factor values replaced by conservation grades or scores. PyMOL can color the structure by B-factor, which allows direct visualization of the conservation pattern.

The score table provides the numerical conservation grades for each residue position. This table is useful for extracting specific values for reporting or for creating custom color schemes. The table includes the residue number, the amino acid type, the conservation grade, and sometimes additional information such as the confidence interval for the score.

### Mapping Scores to Structure Coordinates

The ConSurf output PDB file uses the residue numbering from the structure you provided. Verify that the numbering in the output file matches the numbering in your original structure. If the structure contains multiple chains, confirm that the conservation scores are assigned to the correct chain.

In some cases, the structure may contain residues that are not present in the sequence used for the conservation analysis, or the sequence may contain residues that are not modeled in the structure. These discrepancies should be resolved before visualization. Residues without conservation scores will appear uncolored in PyMOL and should be noted in your interpretation.

### Installing and Configuring PyMOL

PyMOL is a molecular visualization system that is available for Windows, macOS, and Linux. The open-source version provides the core visualization capabilities needed for conservation mapping. The commercial version includes additional features such as the PyMOL Wizard for structure analysis.

After installation, configure PyMOL for conservation visualization by setting the color spectrum that will be used to represent the conservation grades. The standard ConSurf color scheme uses a spectrum from turquoise for variable residues through white for intermediate conservation to burgundy for highly conserved residues. This color scheme can be reproduced in PyMOL using the color commands.

## Visualizing Conservation in PyMOL

### Loading the Structure and Conservation Data

Open PyMOL and load the ConSurf output PDB file using the File menu or the `load` command. The structure will appear in the viewer with the conservation scores stored in the B-factor column. If you prefer to use your original structure file, you can load the conservation scores separately using the `alter` command to replace the B-factor values.

Verify that the structure loaded correctly and that the conservation scores are present. Use the `iterate` command to inspect the B-factor values for a few residues and confirm that they match the scores in the ConSurf output table. This verification step prevents errors from incorrect file mapping.

### Coloring the Structure by Conservation Grade

The most direct way to visualize conservation in PyMOL is to color the structure by B-factor using a spectrum that matches the ConSurf color scheme. The `spectrum` command assigns colors to atoms based on their B-factor values. Specify the color range to match the conservation grade scale from 1 to 9.

For a more nuanced visualization, you can create a custom color spectrum that distinguishes highly conserved, moderately conserved, and variable residues. This approach allows you to emphasize the most conserved regions while still showing the overall conservation pattern. The `color` command with specific B-factor ranges can be used to assign discrete colors to different conservation categories.

### Adjusting the Display for Clarity

The default display in PyMOL shows the protein as a cartoon representation, which is useful for viewing secondary structure but does not show the conservation of individual residues clearly. Switch to a surface or sphere representation to visualize the conservation pattern on the protein surface. The surface representation is particularly useful for identifying conserved patches that may represent binding sites.

For detailed analysis, display the protein as sticks or spheres with the conservation colors. This representation allows you to identify specific conserved residues and examine their side chain orientations. Combine the conservation coloring with other structural features such as secondary structure assignment, ligand positions, or domain boundaries to build a complete picture of the functional regions.

### Creating Publication-Quality Figures

PyMOL can generate high-resolution images suitable for publication. Use the `ray` command to render the image with ray tracing, which produces smooth surfaces and accurate lighting. Set the image size and resolution to match the requirements of your target journal. Add labels to identify important residues, domains, or structural features.

When preparing figures, consider the color scheme for colorblind accessibility. The standard ConSurf color scheme may be difficult to distinguish for some readers. Alternative color schemes that use different hues or patterns can improve accessibility while still conveying the conservation information.

## Interpreting Conservation Scores

### Understanding the Conservation Grade Scale

ConSurf assigns conservation grades from 1 to 9, with 9 being the most conserved. The grades are calculated from the estimated evolutionary rate at each position, with slower rates corresponding to higher conservation. The grades are normalized so that the distribution of scores is approximately uniform across the protein, meaning that a grade of 9 does not indicate absolute conservation but rather that the position is among the most conserved in the protein.

The grade scale is designed for relative comparison within a protein instead of absolute comparison across proteins. A grade of 9 in one protein may represent a different level of sequence identity than a grade of 9 in another protein. When comparing conservation across proteins, use the raw scores or the estimated evolutionary rates instead of the normalized grades.

### Identifying Functionally Important Residues

Residues with conservation grades of 8 or 9 are candidates for functional importance. Examine these residues in the structural context to determine their potential roles. Buried conserved residues may contribute to folding stability or conformational dynamics. Surface conserved residues may participate in binding interactions or catalytic activity. Conserved residues near known functional sites, such as active site residues or ligand binding pockets, are particularly interesting.

Conserved clusters, where multiple highly conserved residues are spatially close in the three-dimensional structure, often indicate functional regions. These clusters may represent catalytic sites, allosteric regulatory sites, or protein-protein interaction interfaces. The spatial clustering of conserved residues provides stronger evidence for functional importance than individual conserved residues in isolation.

### Combining Conservation with Other Evidence

Conservation analysis is most powerful when combined with other types of evidence. Structural information about the local environment, such as solvent accessibility, secondary structure, and proximity to ligands, helps interpret the functional significance of conserved positions. Biophysical predictions about protein dynamics, disorder, and folding can provide additional context. Experimental data from mutagenesis studies, binding assays, or activity measurements can validate or refine the predictions from conservation analysis.

For disease-associated variants, conservation scores can help assess the potential impact of a mutation. A variant at a highly conserved position is more likely to disrupt protein function than a variant at a variable position. However, conservation is only one factor in pathogenicity prediction. Structural context, biochemical properties, and experimental evidence all contribute to the assessment of variant effects.

### Recognizing False Positives and False Negatives

Conservation analysis can produce both false positives and false negatives. False positives are conserved positions that are not actually important for function. These can arise from structural constraints that are not related to function, such as the need to maintain a stable fold, or from evolutionary artifacts such as biased codon usage or alignment errors. False negatives are functionally important positions that show low conservation. These can arise from functional divergence, where different lineages use different mechanisms, or from compensatory mutations that maintain function despite sequence changes.

To reduce false positives and false negatives, examine the alignment quality and the phylogenetic distribution of the sequences. Positions that are conserved only within a subset of the sequences may reflect lineage-specific functions instead of universal importance. Positions that show conservation only among very closely related sequences may not be informative about the broader evolutionary constraints.

## Practical Workflow for Conservation Mapping

### Step-by-Step Protocol

The following protocol outlines the complete workflow from structure selection to final interpretation. Follow these steps in order to ensure reproducible results.

**Step 1: Select the protein structure.** Retrieve the structure from the Protein Data Bank or use a computational model. Record the PDB identifier, chain identifier, resolution or model confidence, and any ligands or modifications present.

**Step 2: Retrieve the protein sequence.** Obtain the amino acid sequence from the NCBI Protein database. Verify that the sequence matches the structure in terms of species, isoform, and residue numbering.

**Step 3: Submit the ConSurf job.** Navigate to the ConSurf web server and submit the sequence. Select the appropriate homolog search and alignment parameters. Provide a valid email address for job notification.

**Step 4: Review the alignment.** When the job is complete, inspect the multiple sequence alignment. Check the number of sequences, the sequence identity distribution, and the alignment quality in regions of interest.

**Step 5: Download the results.** Download the conservation score table and the PDB file with scores in the B-factor column. Save these files with descriptive names that include the protein name and date.

**Step 6: Load the structure in PyMOL.** Open PyMOL and load the ConSurf output PDB file. Verify that the structure and conservation scores loaded correctly.

**Step 7: Color by conservation.** Apply the ConSurf color scheme to the structure. Adjust the display representation to show the conservation pattern clearly.

**Step 8: Identify conserved regions.** Examine the structure for conserved clusters and individual conserved residues. Record the residue numbers and their structural context.

**Step 9: Interpret the results.** Combine the conservation data with structural and functional information to identify candidate functionally important residues. Document your interpretation with specific evidence.

**Step 10: Prepare figures and report.** Create publication-quality figures showing the conservation pattern. Write a summary of the findings, including the methods used and the limitations of the analysis.

### Recording Parameters for Reproducibility

Reproducibility requires careful documentation of all parameters used in the analysis. Record the ConSurf version, the database used for homolog searching, the number of iterations, the expectation value threshold, the maximum number of sequences, the alignment method, and any other settings that could affect the results. Also record the date of the analysis, as database contents change over time and results may differ if the analysis is repeated later.

The ConSurf results page provides a summary of the parameters used for the job. Save this summary along with the output files. If you need to reproduce the analysis or explain the methods in a publication, this information is essential.

### Troubleshooting Common Problems

Several common problems can arise during conservation analysis. If the homolog search returns very few sequences, try increasing the sensitivity of the search or using a different database. If the alignment contains obvious errors, such as misaligned conserved motifs, try a different alignment method or manually curate the alignment. If the conservation scores seem unreasonable, check the sequence used for the analysis and verify that it matches the structure.

If the ConSurf output PDB file does not load correctly in PyMOL, check the file format and the residue numbering. The file may contain nonstandard records that PyMOL does not recognize. In this case, load the original structure and transfer the conservation scores manually using the `alter` command.

## Common Failure Patterns and How to Avoid Them

### Sequence-Structure Mismatch

The most common cause of incorrect conservation mapping is a mismatch between the sequence used for the conservation analysis and the sequence in the structure. This mismatch can arise from using a different isoform, a different species, or a construct with mutations or tags. The result is that conservation scores are assigned to the wrong residues, producing a misleading visualization.

To avoid this problem, always verify that the sequence used for the ConSurf submission matches the sequence in the structure. Compare the residue numbering and the amino acid identities. If the structure contains mutations or tags, decide whether to use the wild-type sequence or the construct sequence and document your choice.

### Poor Alignment Quality

The conservation scores are only as reliable as the multiple sequence alignment. Poor alignments can produce incorrect conservation assignments, particularly in regions with insertions, deletions, or low sequence similarity. The default alignment parameters may not be optimal for all proteins.

To improve alignment quality, inspect the alignment visually and check for obvious errors. Consider using a different alignment method or adjusting the parameters. For proteins with complex evolutionary histories, manual curation of the alignment may be necessary.

### Overinterpretation of Conservation Scores

Conservation scores identify positions under evolutionary constraint but do not directly reveal the functional role of those positions. Interpreting all conserved residues as catalytically important or functionally essential is a common error. Many conserved residues contribute to structural stability or folding instead of to specific biochemical functions.

To avoid overinterpretation, consider the structural context of each conserved residue. Buried residues are more likely to be important for stability, while surface residues are more likely to be involved in interactions. Combine conservation data with other evidence before assigning functional significance.

### Ignoring Alignment Depth and Diversity

Conservation scores from alignments with few sequences or with sequences from closely related species are less reliable than scores from deep and diverse alignments. The ConSurf server provides information about the alignment composition, but researchers sometimes ignore this information and treat all scores as equally reliable.

To avoid this problem, review the alignment statistics before interpreting the results. If the alignment contains fewer than a few dozen sequences or if the sequences are all very similar to the query, consider adjusting the search parameters to obtain a more diverse set of homologs.

## Quality Controls and Validation

### Verifying Conservation Score Consistency

Before relying on the conservation scores, verify that they are internally consistent. Check that residues known to be functionally important from the literature receive high conservation grades. Check that residues in regions known to be variable, such as surface loops, receive low grades. Inconsistencies may indicate problems with the alignment or the scoring.

For proteins with multiple structures available, compare the conservation patterns across the structures. The conservation scores should be similar for the same residues regardless of the structure used, provided the structures represent the same protein. Differences may indicate structural artifacts or errors in the score mapping.

### Cross-Validating with Alternative Methods

Conservation scores from ConSurf can be cross-validated with other computational methods. Sequence-based conservation measures, such as those calculated from multiple sequence alignments using other tools, should correlate with the ConSurf scores. Biophysical feature predictions can provide complementary information about the functional importance of conserved positions.

For example, tools that predict protein dynamics and disorder can indicate whether conserved positions are in rigid or flexible regions. Tools that predict the pathogenicity of missense variants can indicate whether mutations at conserved positions are likely to be deleterious. Agreement between multiple methods strengthens the confidence in the conservation-based predictions.

### Documenting Quality Metrics

Record the quality metrics for the conservation analysis, including the number of sequences in the alignment, the sequence identity range, the alignment method, and the scoring method. These metrics provide context for interpreting the results and allow others to assess the reliability of the analysis. Include these metrics in any report or publication that describes the conservation analysis.

The ConSurf results page provides most of these metrics automatically. Save a copy of the results page or record the metrics in your laboratory notebook. This documentation is essential for reproducibility and for defending the interpretation of the results.

## Limitations and Interpretation Boundaries

### What Conservation Analysis Cannot Tell You

Conservation analysis identifies positions under evolutionary constraint but does not reveal the specific function that is being conserved. A conserved residue could be important for any number of reasons, and additional evidence is required to determine the specific role. Conservation analysis also cannot distinguish between different types of functional importance, such as catalytic activity versus binding affinity versus structural stability.

Conservation analysis is also limited by the availability and quality of homologous sequences. Proteins with few homologs, such as recently evolved proteins or proteins from poorly sampled taxonomic groups, will have less reliable conservation scores. Proteins with complex evolutionary histories, such as those that have undergone domain shuffling or functional divergence, may show conservation patterns that are difficult to interpret.

### Structural Context Limitations

The interpretation of conservation scores depends on the quality and completeness of the protein structure. Regions that are not resolved in the structure, such as flexible loops or disordered termini, cannot be interpreted in structural terms. Low-resolution structures may have inaccurate side chain conformations that affect the assessment of the local environment.

Computational models have additional limitations. The accuracy of the model varies across the structure, and regions with low confidence should be interpreted with caution. Conservation mapping onto a model can still be useful, but the structural interpretation should be tempered by the model quality.

### Evolutionary Context Limitations

The evolutionary signal in conservation scores reflects the history of the protein family. Proteins that have undergone recent functional divergence may show conservation patterns that reflect the new function but not the ancestral function. Proteins that are part of large families with diverse functions may show conservation patterns that reflect the common function of the family instead of the specific function of the query protein.

The choice of homologous sequences also affects the interpretation. Including sequences from a narrow taxonomic range emphasizes recent evolutionary constraints, while including sequences from a broad range emphasizes ancient constraints. The appropriate choice depends on the research question.

## Safety and Ethical Considerations

### Data Handling and Reproducibility

Conservation analysis involves handling sequence and structure data that may be subject to data use agreements or licensing terms. Respect the terms of use for the databases and servers you use. Cite the appropriate databases and tools in any publications that result from the analysis.

Reproducibility is an ethical obligation in scientific research. Document all parameters and procedures so that others can reproduce the analysis. Share the input data, the parameters, and the output files when possible. This transparency allows others to verify the results and to build on the work.

### Responsible Interpretation and Reporting

Conservation analysis results should be reported with appropriate caveats. Avoid overstating the significance of conserved residues without supporting evidence. Report the limitations of the analysis, including the alignment quality, the sequence diversity, and the structural context. This honest reporting allows readers to assess the reliability of the conclusions.

When conservation analysis is used to predict the impact of disease-associated variants, the predictions should be presented as hypotheses instead of definitive conclusions. Conservation is one factor among many that contribute to variant pathogenicity. Clinical decisions should never be based solely on conservation scores.

## Professional Escalation Criteria

### When to Seek Expert Assistance

Conservation analysis is generally accessible to researchers with basic bioinformatics skills, but certain situations warrant consultation with an expert. If the protein has an unusual evolutionary history, such as extensive domain shuffling or horizontal gene transfer, the standard analysis may not be appropriate. If the alignment quality is poor despite parameter adjustments, expert guidance may be needed.

If the conservation results are critical for a high-stakes decision, such as prioritizing mutations for drug development or interpreting clinical variants, consider seeking a second opinion from a structural bioinformatics specialist. Independent verification of the results can identify errors or alternative interpretations.

### When to Use Alternative or Complementary Methods

Conservation analysis is one of many methods for identifying functionally important residues. If the conservation results are ambiguous or inconsistent with other evidence, consider using complementary methods. Molecular dynamics simulations can reveal the dynamic behavior of conserved residues. Mutagenesis studies can experimentally test the functional importance of specific positions. Binding assays can determine whether conserved surface residues participate in interactions.

For proteins with few homologs, alternative methods that do not rely on evolutionary information may be more appropriate. Structure-based methods that identify pockets, cavities, or electrostatic features can complement conservation analysis. Biophysical predictions about protein dynamics and folding can provide additional context.

## Frequently Asked Questions

### What is the minimum number of sequences needed for reliable conservation scores?

The reliability of conservation scores increases with the number and diversity of sequences in the alignment. Alignments with fewer than approximately 20 to 30 sequences may produce scores that are sensitive to the inclusion or exclusion of individual sequences. The ConSurf server provides information about the alignment composition, and you should review this information before interpreting the results. For proteins with very few homologs, consider using complementary methods to identify functionally important residues.

### Can I use conservation analysis on proteins without experimental structures?

Yes, conservation analysis can be performed on computational models. The sequence-based conservation calculation does not require a structure. However, the structural interpretation of the conservation scores requires a three-dimensional model. The reliability of the interpretation depends on the accuracy of the model. Regions with low model confidence should be interpreted with caution, and the limitations of the model should be reported.

### How do I choose between different structures of the same protein?

When multiple structures exist for the same protein, choose the structure that best matches your research question. Consider the resolution for crystallographic structures, the presence of ligands or binding partners, the biological assembly, and the completeness of the modeled regions. The structure with the highest resolution and the most complete coverage is usually the best choice, but the specific research question may favor a different structure.

### What does a conservation grade of 9 mean?

A conservation grade of 9 indicates that the position is among the most conserved in the protein, based on the evolutionary analysis. The grade is normalized so that the distribution of scores is approximately uniform across the protein. A grade of 9 does not necessarily mean that the amino acid is identical across all homologs, but rather that the position evolves more slowly than most other positions in the protein.

### How do I cite ConSurf and PyMOL in my publications?

Cite the original publications for ConSurf and PyMOL, as well as any databases or servers used in the analysis. The ConSurf results page provides citation information for the server. PyMOL should be cited according to the instructions on the PyMOL website. The NCBI databases should be cited according to the instructions on the NCBI website. Check the specific citation requirements of your target journal.

### Can conservation analysis identify all functionally important residues?

No, conservation analysis identifies positions under evolutionary constraint, but not all functionally important residues are conserved. Some functional residues have evolved recently and may not show conservation across diverse species. Some functional roles are maintained by different residues in different lineages, a phenomenon known as functional divergence. Conservation analysis should be combined with other methods to obtain a complete picture of functionally important residues.

### How do I handle proteins with multiple domains?

For proteins with multiple domains, the conservation analysis can be performed on the full-length sequence or on individual domains. Full-length analysis may be appropriate if the domains have similar evolutionary rates. Domain-specific analysis may be more informative if the domains have different evolutionary histories or functions. The choice depends on the research question and the evolutionary context of the protein.

### What is the difference between conservation and sequence identity?

Conservation refers to the evolutionary constraint on a position, which is estimated from the pattern of amino acid substitutions across a phylogeny. Sequence identity refers to the percentage of identical amino acids between two sequences. A position can be conserved even if the amino acid is not identical across all homologs, as long as the substitutions are limited to similar amino acids. Conservation scores account for the biochemical similarity of amino acids, while sequence identity does not.

## 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)
- [Metabolomics Data Analysis in R: A Practical Workflow](/knowledge/bioinformatics/metabolomics-data-analysis-in-r-a-practical-workflow)
- [Microbiome Data Analysis in R: A Practical Guide for Compositional Data](/knowledge/bioinformatics/microbiome-data-analysis-in-r-a-practical-guide-for-compositional-data)
- [Genomic Data Analysis Tools: A Comparative Guide for Researchers](/knowledge/bioinformatics/genomic-data-analysis-tools-a-comparative-guide-for-researchers)
- [Whole Slide Image Analysis: A Practical Workflow for Pathologists](/knowledge/bioinformatics/whole-slide-image-analysis-a-practical-workflow-for-pathologists)

## 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.
- [Rhapsody: predicting the pathogenicity of human missense variants.](https://pubmed.ncbi.nlm.nih.gov/32101277). Bioinformatics (Oxford, England), 2020.
- [b2bTools: online predictions for protein biophysical features and their conservation.](https://pubmed.ncbi.nlm.nih.gov/34057475). Nucleic acids research, 2021.
- [DeepRefiner: high-accuracy protein structure refinement by deep network calibration.](https://pubmed.ncbi.nlm.nih.gov/33999209). Nucleic acids research, 2021.
- [Multi-scale structural analysis of proteins by deep semantic segmentation.](https://pubmed.ncbi.nlm.nih.gov/31424530). Bioinformatics (Oxford, England), 2020.
- [Protlego: a Python package for the analysis and design of chimeric proteins.](https://pubmed.ncbi.nlm.nih.gov/33901273). Bioinformatics (Oxford, England), 2021.

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