Structure analysis
Protein fold recognition
Identify a plausible structural fold for a difficult protein sequence, then review candidate models and templates without hiding the external threading step.
What is protein fold recognition?
Protein fold recognition is a computational method for matching a protein sequence to a known three-dimensional fold even when direct sequence similarity is weak. Threading methods score how well a sequence fits candidate structural templates and return fold or template hypotheses for further modeling and validation.
Use it when profile searches suggest remote homology but do not yield a confident template. Sequence-to-structure compatibility, predicted secondary structure, residue environment, evolutionary profiles, and model-specific energy terms can contribute to ranking.
ProteinIQ does not run a threading engine. Its review workflow starts with an externally generated candidate model and template, then uses FoldSeek and USAlign to inspect structural neighborhood, coverage, TM-score, RMSD, and superposition.
When to use protein fold recognition
- Remote homology is suspected. Use it for difficult template selection when sequence identity is low.
- External threading results are available. Retain the report, candidate model, and selected template.
- Several candidates can be reviewed. Preserve alternatives when methods disagree.
Benefits of protein fold recognition
- Remote structural relationships can be detected. It expands beyond direct sequence matches.
- Templates support difficult models. Candidates can inform comparative-model work.
- Sequence and structure evidence are connected. Review does not discard threading provenance.
Primary limitations
- Missing folds cannot be recovered. The template universe bounds discovery.
- Boundaries and alignments dominate quality. Incorrect correspondence distorts a model.
- A recognized fold does not establish function. Biological claims require further evidence.
Fold-recognition methods and applications
Profile–profile methods compare evolutionary profiles, while threading evaluates sequence compatibility with structural environments or template-derived potentials. Preserve the candidate template, alignment, coverage, score definition, database version, and generated model.
Fold recognition supports remote-homology detection, template selection, domain annotation, and hypothesis generation. It is least reliable for novel folds, disordered proteins, uncertain multidomain boundaries, and states that depend on absent partners or membranes.
How to run protein fold recognition online
- Curate the sequence. Check boundaries, signal peptides, transmembrane regions, low complexity, and domains.
- Run threading externally. Preserve database version, templates, scores, and alignments.
- Select candidates. Retain several plausible templates rather than only the top score.
- Review structures. Load candidate and template into the FoldSeek and USAlign review workflow.
- Validate the hypothesis. Inspect coverage, geometry, conserved residues, sequence evidence, and alternatives.
How to interpret protein fold recognition results
Compare candidates across alignment coverage, structural plausibility, conserved functional positions, geometry, and independent sequence evidence. Do not equate a threading score with experimental structure accuracy or transfer functional annotation without local correspondence and biological support.
How protein fold recognition works
The threading calculation remains external; ProteinIQ reviews its structural candidates rather than recreating the fold-recognition score.
- Curate the sequence. Check sequence boundaries, signal peptides, transmembrane regions, low complexity, and likely domain architecture.
- Run external threading. Run a dedicated fold-recognition or threading method and preserve database version, templates, scores, and alignments.
- Select candidates. Retain several plausible templates rather than selecting only the top score when methods disagree.
- Review structures. Load the candidate model and selected template into the review workflow for FoldSeek and USAlign comparison.
- Validate the hypothesis. Inspect coverage, superposition, geometry, conserved residues, sequence evidence, and alternative folds before downstream use.
Inputs and outputs
Check formats before running, then inspect and download the result from every workflow step.
Inputs
Structure-analysis inputs
PDBmmCIFFASTATSVProtein sequence, external fold-recognition report, candidate model, and selected template structure.
Outputs
Reviewable results
PDBCSVTSVJSONFILESExternal threading provenance plus FoldSeek matches, USAlign scores, residue correspondence, and superposed coordinates.
Tools for protein fold recognition
Use these methods to prepare inputs, run the core analysis, inspect outputs, and validate the evidence described in this workflow.
HMMER
Search profile hidden Markov models for independent sequence-level homology evidence
MMseqs2
Search and cluster large protein sequence collections
FoldSeek
Search structure databases or compare and cluster uploaded protein structures
USAlign
Align two protein structures and return TM-scores, RMSD, residue correspondence, and superposed coordinates
PDB to FASTA converter
Extract protein sequences from coordinate files for sequence-aware review
PDB Download
Retrieve experimental structures from the Protein Data Bank
AlphaFold Database Download
Retrieve predicted protein structures from the AlphaFold Protein Structure Database
DSSP
Assign secondary structure and solvent accessibility from protein coordinates
MolProbity
Check model geometry and steric quality before interpreting structural matches
RMSD calculator
Superpose comparison structures on one reference and report RMSD values
Radius of gyration
Compare global structural compactness across candidate models
Protein parameters
Calculate sequence-derived protein properties for candidate review
Other structure analysis workflows
Compare related approaches based on the molecular system, available evidence, required inputs, and decision you need to support.
Frequently asked questions
Protein sequence, external fold-recognition report, candidate model, and selected template structure.
External threading provenance plus FoldSeek matches, USAlign scores, residue correspondence, and superposed coordinates.
Confirm accession, model, chain, biological assembly, domain boundaries, residue numbering, missing regions, alternate conformations, and prediction confidence. Repair coordinates only when necessary and retain both the original file and every preparation decision.
Use method-native scores together rather than selecting one universal number. TM-score emphasizes length-normalized global fold similarity, RMSD reports geometric deviation over the aligned atoms, and coverage shows how much of each structure actually corresponds.
No. Similar folds can support different functions, and local similarity can occur without shared global architecture. Review residue-level correspondence, domains, ligands, oligomeric state, taxonomy, sequence evidence, curated annotations, and experiments.
A complete protein fold recognition project is generally quote-based. Current providers describe fold recognition and protein-structure analysis as customized services covering data review, method selection, modeling or comparison, validation, and interpretation rather than publishing one universal project price.
The cost depends on structure or sequence count, database scope, model preparation, method comparison, manual inspection, figures, annotation, and whether experimental follow-up is included. Open-source FoldSeek, US-align, and FoldMason can remove a software-license fee, but they do not remove expert analysis or compute requirements.
ProteinIQ self-service starts at $29 per month for academic Plus and $99 per month for commercial Pro, with the configured protein fold recognition workflow estimated in credits before submission. Done-for-you analysis is scoped separately.
Start with a workflow you can inspect and edit
Add your inputs, review the settings, and keep every structure, score, table, and file connected to the step that produced it.