Multiple protein structure alignment

What is multiple protein structure alignment?

Multiple protein structure alignment is the process of placing three or more protein structures into a shared residue correspondence and coordinate frame. The result identifies conserved structural cores, insertions, flexible regions, and fold-level relationships that can remain visible after sequence similarity becomes weak.

Progressive methods such as FoldMason build an alignment from pairwise structural relationships, while other methods optimize a consensus or extend a reference-centered superposition. Structure selection, chains, domains, and conformation strongly affect the result.

ProteinIQ does not run a core multiple-structure alignment engine. Its review workflow accepts externally aligned coordinates, then uses FoldSeek clustering and pyRMSD for separate checks. pyRMSD requires the same selected atoms in the same order across every structure; it cannot infer the residue correspondence created by an external aligner.

When to use multiple protein structure alignment

Use a multiple alignment when the shared structural correspondence is the research question:

• Three or more homologous structures need conserved-core analysis.
• A structural family needs residue-level comparison across members.
• A reviewed external alignment is available for downstream inspection.

Strengths and limitations

An explicit shared mapping can show which residues form a conserved core and which regions vary across a family. FoldSeek clustering provides a broad grouping check; a pairwise RMSD matrix can add geometric detail when all compared structures have identical selected atom correspondence.

Different alignment algorithms can produce different residue mappings. A single reference frame may conceal domain movement, and an RMSD from mismatched atom orders is meaningless even if the matrix calculation completes. Keep chains, domains and conformational states explicit.

How multiple protein structure alignment works

Reference-centered methods align every member to one chosen structure; progressive methods merge pairwise relationships; consensus optimization seeks a mapping shared across the set. Record whether scores describe pairs, columns, or the whole alignment, and retain residue mappings rather than just a mean RMSD.

Applications of multiple protein structure alignment

A shared mapping can locate a catalytic motif or fold core across proteins with weak sequence similarity. The same alignment can guide a structure-informed sequence comparison, provided the residue correspondence and excluded regions remain available for review.

Comparing experimental structures with predicted models can reveal which domains are stable across a family and which vary by state. Domain-wise analysis is often clearer than one global alignment when mobile domains or different assemblies dominate the superposition.

How to run multiple protein structure alignment online

1. Curate structures. Choose comparable chains, assemblies, domains, and conformational states.
2. Run alignment externally. Use a dedicated multiple-structure method and retain its settings, scores, and coordinates.
3. Inspect correspondence. Review cores, gaps, movements, flexible regions, and low coverage.
4. Review structural grouping. Use FoldSeek clustering on the structure set and inspect cluster membership without treating it as a new multiple alignment.
5. Check pairwise geometry where valid. Run pyRMSD only after confirming equal selected atom counts and order in every externally aligned structure.
6. Export provenance. Keep the original residue mapping, aligned coordinates, matrix, exclusions and masks.

How to interpret multiple protein structure alignment results

Read RMSD with aligned residue count and coverage, and verify the atom mapping before interpreting any pyRMSD value. Inspect outliers before removing them: they may be poor models or chains, but can also represent genuine conformational or evolutionary variation. A conserved fold does not confirm shared function or common ancestry by itself.

How multiple protein structure alignment works

The multiple alignment remains external; ProteinIQ reviews the aligned coordinates, clustering, and pairwise RMSD evidence.

1. Curate structures. Choose comparable chains, assemblies, domains, and conformational states, recording experimental or prediction provenance.
2. Run external alignment. Run a dedicated multiple-structure method such as FoldMason and retain its version, parameters, scores, and aligned coordinates.
3. Inspect correspondence. Inspect conserved cores, gaps, domain movements, flexible regions, and structures with unusually low coverage.
4. Review consistency. Load the aligned set into ProteinIQ to compare FoldSeek clustering with the pyRMSD pairwise matrix.
5. Export provenance. Export the original alignment, reviewed coordinates, matrices, exclusions, and any downstream masks or consensus annotations.

Inputs and outputs

Check formats before running, then inspect and download the result from every workflow step.

Inputs

Structure-analysis inputs

PDBmmCIFFASTATSV

Three or more curated protein structures plus aligned coordinates from a dedicated multiple-structure method.

Outputs

Reviewable results

PDBCSVTSVJSONFILES

Multiple-structure correspondence, aligned coordinates, method-native confidence, FoldSeek clusters, and a pyRMSD matrix.