PDB To Fasta¶

Converts one or more PDB structures into FASTA sequences. It parses ATOM records, maps 3-letter residue codes to 1-letter amino acids, and outputs one FASTA record per chain. You can restrict extraction to specific chain IDs or process all chains.

Usage¶

Use this node when you have protein structures (PDB) and need corresponding amino-acid sequences for downstream bioinformatics tasks (e.g., MSA generation, sequence design, or annotation). Typical workflow: load or generate PDB(s) -> PDB To Fasta -> combine or feed the resulting FASTA into alignment or modeling tools.

Inputs¶

Field	Required	Type	Description	Example
pdb	True	PDB	One or more PDB files provided as a structured input. Each entry is treated as a separate structure and parsed for chain sequences.	{'example.pdb': 'ATOM 1 N MET A 1 11.104 13.207 10.118 1.00 20.00 N \nATOM 2 CA MET A 1 12.560 13.332 10.420 1.00 20.00 C \n...'}
chains	True	STRING	Comma-separated list of chain IDs to extract (e.g., "A,B"). Leave empty to extract all chains present in the PDB(s).	A,B

Outputs¶

Field	Type	Description	Example
seqs.fasta	FASTA	FASTA-formatted sequences, one record per chain from each provided PDB. Headers include the PDB entry name and chain ID.	>example.pdb_chain_A MSTNPKPQRIT...

Important Notes¶

Only residues found in ATOM records are parsed; hetero atoms and non-standard residues are not included unless mapped. UNK is mapped to X.
Chain filtering is applied before extraction; leave the 'chains' input empty to process all chains.
If no valid sequence is found for the specified chains, the node raises an error.
Output headers follow the format ">{pdb_name}chain", and sequences are wrapped at 60 characters per line.
Residues are added once per residue number per chain to avoid duplicates from multiple atoms in a single residue.

Troubleshooting¶

No output or error: Ensure the specified chain IDs exist in the PDB (check the chain column). If unsure, leave 'chains' empty to include all chains.
Unexpected amino acids as X: The PDB may contain unknown or unrecognized residue names (UNK) or modified residues that are not mapped.
Very short or empty sequences: Confirm the input is a valid PDB with ATOM records for protein residues; missing ATOM lines or non-protein entries will not yield sequences.
Multiple PDBs input but fewer outputs than expected: Some files may not contain the requested chains; verify chain availability per file.