PDB Fixer¶
Replaces the residue types of a specific chain in one or more PDB structures with a provided FASTA sequence. The node keeps atom coordinates and residue numbering, modifying only the 3-letter residue names to match the FASTA sequence. If the FASTA length differs from the chain length, it replaces as many residues as possible and logs a warning.
Usage¶
Use this node when you need to substitute a placeholder chain (for example, a poly-Gly chain produced by design tools) with a target amino acid sequence while preserving the original PDB geometry. Typical workflow: load or generate a PDB, provide a FASTA sequence for the desired chain, set the target chain ID (e.g., 'B'), and feed the fixed PDB into downstream modeling or design nodes.
Inputs¶
| Field | Required | Type | Description | Example |
|---|---|---|---|---|
| pdb | True | PDB | PDB structure(s) to modify. Expects a dictionary mapping PDB IDs to PDB text content. | {'template_1': 'ATOM 1 N GLY A 1 ...', 'template_2': 'ATOM 1 N GLY B 1 ...'} |
| fasta | True | FASTA | FASTA sequence to apply to the target chain. Headers are ignored; only sequence lines are used. Whitespace is stripped. | >target_chain_B MKTFFVAGL |
| target_chain | True | STRING | Single-character chain ID to replace (e.g., 'A', 'B'). | B |
Outputs¶
| Field | Type | Description | Example |
|---|---|---|---|
| fixed_pdb | PDB | Dictionary of PDB structures with the target chain's residue names replaced according to the FASTA sequence while retaining coordinates and numbering. | {'template_1': 'ATOM 1 N MET B 1 ...', 'template_2': 'ATOM 1 N LYS B 1 ...'} |
Important Notes¶
- Input format: The PDB input must be a non-empty dictionary {pdb_name: pdb_string}.
- Target chain: Must be a single alphanumeric character matching the chain ID present in the PDB (position 22 in standard PDB lines).
- FASTA parsing: Headers (lines starting with '>') are ignored; sequence lines are concatenated with whitespace removed.
- Residue mapping: The node maps 1-letter amino acids to 3-letter codes (A→ALA, R→ARG, ..., U→SEC, O→PYL, X→UNK). Unknown letters are left as original residue names.
- Length mismatch: If the FASTA length differs from the number of residues detected in the target chain, it replaces residues up to the shorter length and logs a warning.
- Scope of changes: Only residue names are changed for ATOM/HETATM records of the target chain; coordinates, atom names, residue numbering, and all other records remain unchanged.
- Per-residue logic: Replacement is based on residue sequence order inferred from residue number changes (unique residues by number). Multiple atoms for the same residue are updated consistently.
- Multiple PDBs: Supports batch operation; each PDB entry in the input dictionary is processed independently.
- Requirements: PDB lines must follow standard fixed-column formatting for chain ID (col 22) and residue number (cols 23–26).
- Failure cases: The node raises an error if the PDB dict is empty/invalid, the chain is not found, or the FASTA contains no sequence.
Troubleshooting¶
- Chain not found: Error like "Chain 'B' not found" indicates the target_chain does not exist in the input PDB. Verify the chain ID and that your PDB uses standard chain identifiers.
- Empty or invalid FASTA: If you see "FASTA content cannot be empty" or "No sequence found in FASTA data", ensure your FASTA includes sequence lines beneath any header.
- Length mismatch warning: If you get a warning about different lengths, confirm your FASTA matches the chain length. The node will still replace up to the shorter length.
- Unknown amino acids: If warnings appear about unknown letters, replace nonstandard characters in the FASTA or note that those residues will keep original 3-letter names.
- Unexpected output structure: If the output is not usable downstream, ensure the input was a dict of {id: pdb_content} and that your PDB lines maintain correct column formatting.