similar-protein-retrieval

To find protein homologs, provide a UniProt ID, PDB ID, FASTA string, or PDB file as input. The Skill parses the identifier and runs sequence-based MSA or structure-based FoldSeek workflows to retrieve similar proteins.

Sequence-based protein similarity search discovers homologs via MSA results, while structure-based search identifies proteins with highly similar folds using FoldSeek. Both approaches support functional inference and cross-species comparisons.

Yes, you can input a FASTA string to run similarity searches for cross-species comparisons. The Skill integrates FoldSeek for structure-based search and MSA workflows for sequence-based homolog discovery.

The best way to retrieve proteins with similar folds is using the structure-based search capability powered by FoldSeek. It identifies proteins with highly similar structures to your query for functional annotation.

No, you do not need a PDB file exclusively. You can run a protein structure similarity search by providing a UniProt ID, PDB ID, or FASTA string, which the Skill parses to integrate with FoldSeek workflows.

Robust error handling ensures that invalid or unsupported protein identifiers are caught during input parsing. This prevents workflow failures when processing UniProt IDs, PDB IDs, FASTA strings, or PDB files for homolog discovery.