def validate_fasta_file(file_content, session_dir, file_name):
if file_content is None or file_name is None or session_dir is None:
return None
else:
if ".zip" in file_name:
file_content = tools.decode_zip_content(file_content)
else:
file_content = tools.decode_content(file_content)
output_dir = Path(session_dir)
fasta_path = tools.get_child_path(output_dir, file_name)
if ".zip" in file_name:
tools.save_to_file(file_content, fasta_path, 'wb')
else:
tools.save_to_file(file_content, fasta_path)
error_message = pangtreebuild.fasta_file_is_valid(fasta_path)
if len(error_message) == 0:
return {
"is_correct": True,
"filename": file_name,
"error": error_message
}
else:
return {
"is_correct": False,
"filename": file_name,
"error": error_message
}
def get_elem_cache_info(pangenome_hash: int) -> Path:
cache_file_name = str(abs(pangenome_hash)) + ".pickle"
parent_output_dir = Path(
os.path.abspath(os.path.join(
os.path.dirname(__file__)))).joinpath("../cache/").resolve()
poagraph_elements_cache_path = tools.get_child_path(
parent_output_dir, cache_file_name)
return poagraph_elements_cache_path
def run_pangtreebuild(output_dir: Path,
datatype: DataType,
multialignment: Union[Maf, Po],
fasta_provider: Union[FromFile, FromNCBI,
ConstBaseProvider],
blosum: Blosum,
consensus_choice: str,
output_po: bool,
output_fasta: bool,
output_newick: bool,
missing_symbol: MissingBase,
metadata: Optional[MetadataCSV] = None,
hbmin: Optional[Hbmin] = None,
stop: Optional[Stop] = None,
p: Optional[P] = None,
fasta_path: Optional[Path] = None,
include_nodes: Optional[bool] = None) -> PangenomeJSON:
start = time.time()
logprocess.add_file_handler_to_logger(output_dir,
"details",
"details.log",
propagate=False)
logprocess.add_file_handler_to_logger(output_dir,
"",
"details.log",
propagate=False)
logprocess.remove_console_handler_from_root_logger()
poagraph, dagmaf = None, None
if isinstance(multialignment, Maf):
poagraph, dagmaf = builder.build_from_dagmaf(multialignment,
fasta_provider, metadata)
elif isinstance(multialignment, Po):
poagraph = builder.build_from_po(multialignment, metadata)
consensus_output_dir = tools.get_child_dir(output_dir, "consensus")
consensus_tree = None
if consensus_choice == 'poa':
consensus_tree = build_poa_affinity_tree(poagraph, blosum,
consensus_output_dir, hbmin,
True)
elif consensus_choice == 'tree':
consensus_tree = build_affinity_tree(poagraph, blosum,
consensus_output_dir, stop, p,
True)
if output_po:
pangenome_po = poagraph_to_PangenomePO(poagraph)
tools.save_to_file(pangenome_po,
tools.get_child_path(output_dir, "poagraph.po"))
if output_fasta:
sequences_fasta = poagraph_to_fasta(poagraph)
tools.save_to_file(sequences_fasta,
tools.get_child_path(output_dir, "sequences.fasta"))
if consensus_tree:
consensuses_fasta = affinity_tree_to_fasta(poagraph,
consensus_tree)
tools.save_to_file(
consensuses_fasta,
tools.get_child_path(output_dir, "consensuses.fasta"))
if output_newick:
if metadata is not None:
seq_id_to_metadata = {
seq_id: seq.seqmetadata
for seq_id, seq in poagraph.sequences.items()
}
else:
seq_id_to_metadata = None
affinity_tree_newick = consensus_tree.as_newick(seq_id_to_metadata,
separate_leaves=True)
tools.save_to_file(
affinity_tree_newick,
tools.get_child_path(output_dir, "affinity_tree.newick"))
end = time.time()
task_parameters = TaskParameters(
running_time=f"{end - start}s",
multialignment_file_path=multialignment.filename,
multialignment_format=str(type(multialignment).__name__),
datatype=datatype.name,
metadata_file_path=metadata.filename if metadata else None,
blosum_file_path=blosum.filepath.name,
output_path=None,
output_po=output_po,
output_fasta=output_fasta,
output_with_nodes=include_nodes,
verbose=True,
raw_maf=False,
fasta_provider=str(type(fasta_provider).__name__),
missing_base_symbol=missing_symbol.value,
fasta_source_file=fasta_path,
consensus_type=consensus_choice,
hbmin=hbmin.value if hbmin else None,
stop=stop.value if stop else None,
p=p.value if p else None)
pangenomejson = to_PangenomeJSON(task_parameters=task_parameters,
poagraph=poagraph,
dagmaf=dagmaf,
affinity_tree=consensus_tree)
pangenome_json_str = to_json(pangenomejson)
tools.save_to_file(pangenome_json_str,
tools.get_child_path(output_dir, "pangenome.json"))
return pangenomejson
def get_default_blosum_path():
parent_dir = Path(os.path.dirname(os.path.abspath(__file__)) + '/')
return tools.get_child_path(parent_dir, "../dependencies/blosum80.mat")
def run_poapangenome(output_dir: Path,
datatype: DataType,
multialignment: Union[Maf, Po],
fasta_provider: Union[FromFile, FromNCBI,
ConstSymbolProvider],
blosum: Blosum,
consensus_choice: str,
output_po: bool,
output_fasta: bool,
missing_symbol: MissingSymbol,
metadata: Optional[MetadataCSV] = None,
hbmin: Optional[Hbmin] = None,
stop: Optional[Stop] = None,
p: Optional[P] = None,
fasta_path: Optional[Path] = None) -> PangenomeJSON:
start = time.time()
logprocess.add_file_handler_to_logger(output_dir,
"details",
"details.log",
propagate=False)
logprocess.add_file_handler_to_logger(output_dir,
"",
"details.log",
propagate=False)
logprocess.remove_console_handler_from_root_logger()
poagraph, dagmaf = None, None
if isinstance(multialignment, Maf):
poagraph, dagmaf = Poagraph.build_from_dagmaf(multialignment,
fasta_provider, metadata)
elif isinstance(multialignment, Po):
poagraph = Poagraph.build_from_po(multialignment, metadata)
consensus_output_dir = tools.get_child_dir(output_dir, "consensus")
consensus_tree = None
if consensus_choice == 'poa':
consensus_tree = simple_tree_generator.get_simple_consensus_tree(
poagraph, blosum, consensus_output_dir, hbmin, True)
elif consensus_choice == 'tree':
consensus_tree = tree_generator.get_consensus_tree(
poagraph, blosum, consensus_output_dir, stop, p, MAX2(), NODE3(),
True)
if output_po:
pangenome_po = poagraph_to_PangenomePO(poagraph)
tools.save_to_file(pangenome_po,
tools.get_child_path(output_dir, "poagraph.po"))
if output_fasta:
sequences_fasta = poagraph_to_fasta(poagraph)
tools.save_to_file(sequences_fasta,
tools.get_child_path(output_dir, "sequences.fasta"))
if consensus_tree:
consensuses_fasta = consensuses_tree_to_fasta(
poagraph, consensus_tree)
tools.save_to_file(
consensuses_fasta,
tools.get_child_path(output_dir, "consensuses.fasta"))
end = time.time()
task_parameters = TaskParameters(
running_time=f"{end - start}s",
multialignment_file_path=multialignment.filename,
multialignment_format=str(type(multialignment).__name__),
datatype=datatype.name,
metadata_file_path=metadata.filename if metadata else None,
blosum_file_path=blosum.filepath.name,
output_path=None,
output_po=output_po,
output_fasta=output_fasta,
output_with_nodes=True,
verbose=True,
raw_maf=False,
fasta_provider=str(type(fasta_provider).__name__),
missing_base_symbol=missing_symbol.value,
fasta_source_file=fasta_path,
consensus_type=consensus_choice,
hbmin=hbmin.value if hbmin else None,
max_cutoff_option="MAX2",
search_range=None,
node_cutoff_option="NODE3",
multiplier=None,
stop=stop.value if stop else None,
p=p.value if p else None)
pangenomejson = to_PangenomeJSON(task_parameters=task_parameters,
poagraph=poagraph,
dagmaf=dagmaf,
consensuses_tree=consensus_tree)
pangenome_json_str = to_json(pangenomejson)
tools.save_to_file(pangenome_json_str,
tools.get_child_path(output_dir, "pangenome.json"))
return pangenomejson
def run_pangenome(run_processing_btn_click, session_state: Dict,
session_dir: str, datatype: str, multialignment_content: str,
multialignment_filename: str, fasta_provider_choice: str,
fasta_content: str, fasta_filename: str, missing_symbol: str,
consensus_choice: str, output_config: List[str],
metadata_content: str, metadata_filename: str,
hbmin_value: float, stop_value: float, p_value: float):
if run_processing_btn_click == 0:
raise PreventUpdate()
if session_state is None:
session_state = {}
if session_dir is None:
session_dir = tools.create_output_dir()
else:
session_dir = Path(session_dir)
current_processing_output_dir_name = tools.get_child_path(
session_dir, tools.get_current_time())
tools.create_dir(current_processing_output_dir_name)
if multialignment_filename and "maf" in multialignment_filename:
multialignment = Maf(StringIO(
tools.decode_content(multialignment_content)),
file_name=multialignment_filename)
elif multialignment_filename and "po" in multialignment_filename:
multialignment = Po(StringIO(
tools.decode_content(multialignment_content)),
file_name=multialignment_filename)
else:
session_state[
"error"] = "Cannot create Poagraph. Only MAF and PO files are supported."
return session_state
missing_symbol = MissingBase(
missing_symbol) if missing_symbol != "" else MissingBase()
fasta_path = None
if fasta_provider_choice == "NCBI":
fasta_provider = FromNCBI(use_cache=True)
elif fasta_provider_choice == "File":
fasta_path = tools.get_child_path(current_processing_output_dir_name,
fasta_filename).resolve()
save_mode = "wb" if "zip" in fasta_filename else "w"
if "zip" in fasta_filename:
fasta_decoded_content = tools.decode_zip_content(fasta_content)
else:
fasta_decoded_content = tools.decode_content(fasta_content)
tools.save_to_file(fasta_decoded_content, fasta_path, save_mode)
fasta_provider = FromFile(fasta_path)
blosum_path = pangtreebuild.get_default_blosum_path()
blosum_contents = tools.read_file_to_stream(blosum_path)
blosum = Blosum(blosum_contents, blosum_path)
metadata = MetadataCSV(StringIO(tools.decode_content(metadata_content)),
metadata_filename) if metadata_content else None
pangenomejson = pangtreebuild.run_pangtreebuild(
output_dir=current_processing_output_dir_name,
datatype=DataType[datatype],
multialignment=multialignment,
fasta_provider=fasta_provider,
blosum=blosum,
consensus_choice=consensus_choice,
output_po=True if "po" in output_config else False,
output_fasta=True if "fasta" in output_config else False,
output_newick=True if "newick" in output_config else False,
missing_symbol=missing_symbol,
metadata=metadata,
hbmin=Hbmin(hbmin_value) if hbmin_value else None,
stop=Stop(stop_value) if stop_value else None,
p=P(p_value) if p_value else None,
fasta_path=fasta_filename if fasta_filename else None,
include_nodes=True # if "nodes" in output_config else False
)
pangenome_json_str = to_json(pangenomejson)
current_processing_output_zip = tools.dir_to_zip(
current_processing_output_dir_name)
current_processing_short_name = "/".join(
str(current_processing_output_zip).split("/")[-2:])
return {
"last_output_zip": current_processing_short_name,
"jsonpangenome": pangenome_json_str,
"error": ""
}, True