automatic official FASTA file fetching, several new utility functions related to structure including flexible 3d alignment that supports different length chains to be aligned! #101
Conversation
…he clusters centers/representatives including the sequence. this is useful for create splits etc.
…ng few parts in structure_io
| from typing import Optional | ||
|
|
||
|
|
||
| def main( |
There was a problem hiding this comment.
what is this script for? please explain in the comments section.
There was a problem hiding this comment.
added this desc in the docstring:
"Takes an input PDB files and splits it into separate files, one per describe chain, allowing to rename the chains if desired"
| mask=None, # TODO: check | ||
| ) | ||
|
|
||
| # apply_on_atom_pos = apply_rigid_on_dynamic_concat['atom_positions'] |
There was a problem hiding this comment.
Do you need the commented out code below?
| @@ -22,18 +22,26 @@ def __init__(self, verbose: bool = True) -> None: | |||
| self.chains_data = {} # maps from chain description (e.g. ('7vux', 'A')) to | |||
| self.flattened_data = {} | |||
|
|
|||
| # | |||
| self.per_chain_most_frequent_residue_part = {} | |||
There was a problem hiding this comment.
a key is a tuple in the format (pdb_id, chain_id)
added a comment above it with description
| def add( | ||
| self, | ||
| pdb_id: str, | ||
| pdb_id_or_filename: str, | ||
| pdb_id: Optional[str] = None, |
There was a problem hiding this comment.
Why do you keep pdb_id? Is backward compatibility important here?
There was a problem hiding this comment.
pdb_id_or_filename might be a path to a local pdb file, so pdb_id gives the user an option to define the pdb_id as there is no implemented way to automatically extract it.
I'll check ways to extract it automatically from the pdb file itself, and check if it's important to allow the user to override that.
That will be a separate PR.
about what you asked - backward compatibility isn't super critical here, as we can likely detect all usage locations and port them to newer usage.
| @@ -174,7 +176,7 @@ def load_protein_structure_features( | |||
| chain_id_type: str = "author_assigned", | |||
| device: str = "cpu", | |||
| max_allowed_file_size_mbs: float = None, | |||
| also_return_mmcif_object: bool = False, | |||
| # also_return_mmcif_object: bool = False, | |||
| ): | ||
| if not m_res: | ||
| continue | ||
| aa_idx = aa_idx.item() | ||
| p_res = p_res.clone().detach().cpu() # fixme: this looks slow | ||
| if aa_idx == 21: | ||
| # if torch.is_tensor(p_res): |
No description provided.