def seq_guess_and_write(seqs, filename):
seq_format, compressed = utilities.guessFileFormat(filename)
if seq_format is not None:
with utilities.flexible_handle(filename, compressed, 'wt') as seq_out:
SeqIO.write(seqs, seq_out, seq_format)
else:
print("Cannot infer sequence format for file: " + filename)
def cleanupAndWrite(assembly_file,output_file,length=None,coverage=None,image_file=None,base_ID=None):
##Note: no sanity checks
## Load the assemblies
assembly_format,assembly_compressed = utilities.guessFileFormat(assembly_file)
output_format,output_compressed = utilities.guessFileFormat(output_file)
if assembly_format != output_format:
print("Warning on cleanup: input and output formats do not match ({} and {})".format(assembly_format,output_format))
with utilities.flexible_handle(assembly_file, assembly_compressed, 'rt') as fin:
seqs = [c for c in SeqIO.parse(fin,assembly_format)]
if base_ID is not None:
new_contigs, c = seq_utilities.standardize_contig_names(seqs,base_ID)
seqs = new_contigs
#Precise manipulation of single contig
if length is None:
length = 0
if coverage is None:
coverage = 0
##always SPADES
print("Removing low quality contigs from SKESA assembly. Length < {}; coverage < {}".format(length,coverage))
# raw_filename = os.path.join(os.path.dirname(report_file),os.path.basename(assembly_file))
discard_file = utilities.appendToFilename(output_file, '_discarded') ##ext is same as assembly file
updated_seqs = cleanup_SKESA(seqs,minimum_length = length, minimum_coverage = coverage,discard_file=discard_file,export_contig_graph=image_file)
if updated_seqs is None:
print("Unable to clean and orient the assembly: \n\t"+assembly_file)
return 1
else:
print("Retained {} of {} contigs.".format(len(updated_seqs),len(seqs)))
with open(output_file,'wt') as fout:
SeqIO.write(updated_seqs,fout,output_format)
print('Saved reoriented assembly at {}'.format(output_file))
if output_compressed:
print("Warning. Compression not implemented. The file extension is misleading")
return 0
def seqs_guess_and_parse2list(filename):
seq = None
seq_format, compressed = utilities.guessFileFormat(filename)
if seq_format is not None:
with utilities.flexible_handle(filename, compressed, 'rt') as seq_in:
seq = [x for x in SeqIO.parse(seq_in, seq_format)]
else:
print("Cannot infer sequence format for file: " + filename)
return seq
def seqs_guess_and_parse2dict(filename):
if not isinstance(filename, str):
raise TypeError("Filename must be string, is {}".format(
type(filename)))
# if not os.path.isfile(filename):
# raise ValueError("Cannot locate file: {}".format(filename))
seq_dict = None
seq_format, compressed = utilities.guessFileFormat(filename)
if seq_format is not None:
with utilities.flexible_handle(filename, compressed, 'rt') as seq_in:
seq_dict = SeqIO.to_dict(SeqIO.parse(seq_in, seq_format))
else:
print("Cannot infer sequence format for file: " + filename)
return seq_dict
def describeSequences(sequenceFile):
result = defaultdict(int)
result['FileSize'] = os.path.getsize(sequenceFile)
seq_format, compressed = utilities.guessFileFormat(
sequenceFile) ##guess and parse
if seq_format is not None:
with utilities.flexible_handle(sequenceFile, compressed,
'rt') as seq_in:
for s in SeqIO.parse(seq_in, seq_format):
result['Sequences'] += 1
result['Nucleotides'] += len(s)
else:
print("Cannot infer sequence format for file: " + sequenceFile)
##TODO: add Q30 and such?
return result
def setupGenomeForBlastBasedExtraction(genome_name,genome_file,tempDir,file_format = '',is_compressed = None):
##Genome information
genomeInfo = dict()
genomeInfo['name'] = genome_name
genomeInfo['original'] = genome_file #just for reporting
##Some people use weird genome filenames, so I need to copy it to something without special characters
temp_genome = os.path.join(tempDir,genome_name + '.fasta')
genomeOrganizer.exportGenomeFASTA(genome_file,temp_genome,file_format,is_compressed)
genomeInfo['fasta'] = temp_genome
if not os.path.isfile(genomeInfo['fasta']):
raise IOError("Illegitimate file at "+genomeInfo['fasta'])
#~ genomeDir,genomeFile = os.path.split(os.path.abspath(genomeInfo['fasta']))
#open the genome file for extracting sequences
genome_handle = utilities.flexible_handle(genomeInfo['original'], is_compressed, 'rt')
genomeInfo['seqs'] = SeqIO.to_dict(SeqIO.parse(genome_handle, file_format))
print("{} bp in {} contig(s)".format(sum([len(c) for c in genomeInfo['seqs'].values()]),len(genomeInfo['seqs']))) ##Appends to sequence identifier line
if len(genomeInfo['seqs']) == 0:
raise ValueError("No sequences parsed from file {}".format(genomeInfo['fasta']))
genome_handle.close()
# make search database for genome
db_base = os.path.basename(genomeInfo['fasta'])
genomeInfo['db'] = os.path.join(tempDir,db_base)
makeblastdb(genomeInfo['fasta'],genomeInfo['db'])
return genomeInfo
def cleanupAndWrite(assembly_file,
output_file,
circle_new_start=None,
reverse_contig=None,
closed_circle=None,
broken_circle=None,
circularize_with_Ns=0,
length=None,
coverage=None,
report_file=None,
reference=None,
assembler=None,
working_dir=None):
##Note: no sanity checks
## Load the assemblies
assembly_format, assembly_compressed = utilities.guessFileFormat(
assembly_file)
output_format, output_compressed = utilities.guessFileFormat(output_file)
if assembly_format != output_format:
print(
"Warning on cleanup: input and output formats do not match ({} and {})"
.format(assembly_format, output_format))
with utilities.flexible_handle(assembly_file, assembly_compressed,
'rt') as fin:
seqs = [c for c in SeqIO.parse(fin, assembly_format)]
#Precise manipulation of single contig
updated_seqs = None
if circle_new_start or reverse_contig:
if len(seqs) > 1:
print(
"Error: User provided explicit reorientation instructions for a contig, but multiple contigs are present in assembly: \n"
+ assembly_file)
return 1
elif closed_circle:
print("Shifting closed circle...")
updated_seqs = shiftCirclarChromosome(seqs[0],
circle_new_start,
reverse_contig,
N_padding=0)
elif broken_circle:
print("Shifting broken circle...")
updated_seqs = shiftCirclarChromosome(seqs[0],
circle_new_start,
reverse_contig,
N_padding=-1)
elif circularize_with_Ns > 0:
print('Scaffolding not implemented')
else:
print(
'To shift a chromosome, you must specify whether the circle is closed or broken'
)
else: ## Complex criteria for manipulation
if closed_circle and len(seqs) > 1:
print(
"Warning: Untested parameters. User specified 'closed circle' but multiple contigs are present in assembly"
)
## Remove the low-quality contigs:
##TODO: consider if another parameter should be passed. At least specify if it came from SPAdes
circular = closed_circle or broken_circle ##Circles imply high-quality sequence
if not circular:
if length is None:
length = 0
if coverage is None:
coverage = 0
if assembler is None:
print("Removing short contigs from assembly.")
updated_seqs = [x for x in seqs if len(x) > length]
# if coverage
elif assembler.upper() == 'SPADES':
print(
"Removing low quality contigs from SPADES assembly. Length < {}; coverage < {}"
.format(length, coverage))
raw_filename = os.path.join(os.path.dirname(report_file),
os.path.basename(assembly_file))
image_file = None # utilities.setExt(raw_filename, 'png') ##Note: this has been moved to the calculateStats routine
discard_file = utilities.appendToFilename(
raw_filename, '_discarded') ##ext is same as assembly file
updated_seqs = cleanup_SPADES(seqs,
minimum_length=length,
minimum_coverage=coverage,
export_contig_data=report_file,
discard_file=discard_file,
export_contig_graph=image_file)
else:
print(
"Error: assembler ({}) unknown for non-circular assembly. Not attempting to cleanup contigs in file: \n{}"
.format(assembler, assembly_file))
return 1
## Reorient to reference if requested
if reference:
input_seqs = updated_seqs if updated_seqs is not None else seqs
if os.path.isfile(reference):
if circular:
if len(input_seqs) > 1:
print(
'Warning: multiple contigs in "circular" assembly. Only one contig will be reoriented and I cannot tell you which one. Untested.'
)
if len(input_seqs) > 0:
N_padding = -1 ##Do not religate
if closed_circle:
N_padding = 0
elif circularize_with_Ns > 0:
print('Scaffolding not implemented')
return 1
print(
"Reorienting circular chromosome to reference...")
updated_seqs = reorientClosedChromosome(
input_seqs,
reference,
N_padding=N_padding,
working_dir=working_dir
) #Note: only treated as closed if N_padding >= 0
else: ## Len == 0
print(
"None of {} contigs passed your exclusion criteria. Exiting "
.format(len(seqs)))
return 1
else:
if working_dir is None:
working_dir = os.path.splitext(output_file)[0]
draft_name = os.path.splitext(
os.path.basename(assembly_file))[0]
print("Reorienting contigs")
reorder_stats = reorientContigs(
input_seqs,
reference,
working_dir,
name=draft_name,
input_format=assembly_format) ##Will be genbank format
if isinstance(reorder_stats, dict) and ('ReorderedDraft'
in reorder_stats):
updated_seqs = seq_utilities.seqs_guess_and_parse2list(
reorder_stats['ReorderedDraft']
) ##Excessive to reload... but it fits in this flow
else:
updated_seqs = None
else:
print(
"Unable to realign to reference because there is no refernce file: {}"
.format(reference))
if updated_seqs is None:
print("Unable to clean and orient the assembly: \n\t" + assembly_file)
return 1
else:
with open(output_file, 'wt') as fout:
SeqIO.write(updated_seqs, fout, output_format)
print('Saved cleaned assembly at {}'.format(output_file))
if output_compressed:
print(
"Warning. Compression not implemented. The file extension is misleading"
)
return 0