This is a tool to compute the ordering or nucleotides in a PDB file.
Put PDB files in tmp/
and run make
. This will produce a out/resolved.json
which will contain all data processed by bin/resolved.py
.
Note that many structures are discontinuous, which will cause BioPython to spit out lots of warnings.
-
requirements.txt
A pip requirements file for this.
-
in/known.json
This is a json file of all known residue types. This is used to filter results from reading a PDB to only extract known residues. Many chains have a lot of water and ions or ligands in them. We wish to skip these so anything not found in this file is skipped. Currently it only contains RNA nucleotides and modified nucleotides.
The format is a simple modified-type -> standard-type object. This mapping is used to determine the sequence of each chain. For example
2MA
which is a modifiedA
will be represented as anA
in the sequence. While the nucleotide ids will use the correct2MA
unit.
This is a script to parse one or more PDB files and determine the ordering of nucleotides, the resolved sequence and the nucleotide ids. This will produce one JSON object written to standard out of the form:
{"pdb_id": {"ordering": [nt1, nt2, ...]
chain_id: {"sequence": sequence_string, "residues": [nt1, nt2] }
}
...
}
This assumes that each file is named pdb_id.pdb. The ordering entry is the global ordering of all nucleotides in the file, while the residue entries for each object are the residues in order for that chain.
Currently this does not determine asymmetric unit/biological assembly correctly, it always assumes asymmetric unit.
Blake Sweeney bsweene@bgsu.edu