Structure & Sequence File Prep
Project Tools (VenusFactory2)
| Tool | Purpose |
|---|---|
| read_fasta | Parse multi-FASTA → headers/sequences |
| extract_uids_from_fasta | Pull UIDs from headers |
| uid_file_to_chunks | Chunk UID lists for batch jobs |
| pdb_chain_sequences | Per-chain sequences from PDB |
| get_seq_from_pdb_chain_a | Chain A sequence shortcut |
| pdb_dir_to_fasta | Directory of PDBs → FASTA |
| check_pdb_apo | Ligand-free / apo heuristic |
| maxit_structure_convert | pdb2cif / cif2pdb / cif2mmcif (needs MAXIT) |
| extract_uniprot_id_from_rcsb_metadata | UniProt from RCSB metadata JSON |
| unzip_archive / ungzip_file | Unpack downloads |
Workflows
PDB → design/mutation ready
check_pdb_apo(if apo backbone required)pdb_chain_sequences→ choose designed chains- Optional
maxit_structure_convertfor format mismatches
RCSB metadata → UniProt → AlphaFold
download_rcsb_entry_metadata_by_pdb_idextract_uniprot_id_from_rcsb_metadata- Hand off to
alphafold_database/protein_structure_pipeline
Common mistakes
- Hardcoding chain lengths without
pdb_chain_sequences - Calling MAXIT strategies other than
pdb2cif/cif2pdb/cif2mmcif - Feeding directories to tools that expect a single file path