def blastn(self, outputdir):
# create a dir for output
create_outputdir(outputdir)
# blastn can only run with fasta files, so input has to be converted
if is_fastq(self.input):
# print actual informations about the step on stdout
print_step(self.step_number,
'Annotation',
'convert fastq files',
cut_path(self.input))
newline()
self.input = convert_fastq(self.input, self.blast_dir, self.converter_exe)
# blastn can only annotated one file, so input has to be merged to one file
if is_paired(self.input):
# print actual informations about the step on stdout
print_step(self.step_number,
'Annotation',
'merging reads to on file',
cut_path(self.input))
newline()
self.input = merge_files(self.input, self.blast_dir, 'merged', 'fasta')
# define the outputformat for the blastn results
outfile = outputdir + os.sep + blast_output(self.outfmt)
# print actual informations about the step on stdout
print_step(self.step_number,
'Annotation',
'blast sequences against nt database',
self.blast_parameter)
newline()
# start blastn and wait until completion
# logfile is not requiered, because blastn has no log function and no output to stdout
p = subprocess.Popen(shlex.split('%s -db %s -query %s -out %s -num_threads %s %s ' %
(self.blastn_exe,
self.blastn_db,
to_string(self.input),
outfile,
self.threads,
self.blast_parameter)),
stderr = open_logfile(self.logdir + 'blastn.err.log'))
# wait until process is complete
p.wait()
if p.returncode:
raise BlastnException(self.logdir + 'blastn.err.log')
else:
# remove the temporary files: converted fastq files and the merged fasta files
remove_file(outputdir + os.sep, 'converted', 'fasta')
remove_file(outputdir + os.sep, 'merged', 'fasta')
# remove unused error logs
remove_empty_logfile(self.logdir + 'blastn.err.log')
# print summary of the process after completion
print_verbose('Annotation with blastn complete \n')
print_running_time(self.time)
newline()
def manage_assembly(self):
concatinated = ''
assembled = ''
# run the assembling functions when the module is initialized
if self.mode.lower() == 'flash':
# is executable existing and runnable?
if is_executable(self.flash_exe):
# start concatination and update the input for next step
self.concatinate(self.concat_out)
self.step_number += 1
# merge the concatinated reads with non concatinated rest
if (self.merge_uncombined):
self.input = merge_files([to_string(update_reads(self.concat_out, 'extendedFrags', 'fastq')),
to_string(update_reads(self.concat_out, 'out.notCombined', 'fastq'))],
self.concat_out,
'merged_concat','fastq')
else:
concatinated = update_reads(self.concat_out, 'extendedFrags', 'fastq')
self.input = concatinated
if self.mode.lower() == 'metavelvet':
# is executable existing and runnable?
if is_executable(self.velveth_exe) and is_executable(self.velveth_exe) and is_executable(self.metavelvet_exe):
# start assembly and update the input for next step
self.assemble_reads(self.assembly_out)
assembled = update_reads(self.assembly_out, 'meta-velvetg', 'fa')
self.input = assembled
self.step_number += 1
if self.mode.lower() == 'both':
#TODO: not working because of auto mode --> see logs
# is executable existing and runnable?
if is_executable(self.flash_exe) and is_executable(self.velveth_exe) and is_executable(self.velveth_exe) and is_executable(self.metavelvet_exe):
# start processing and update the input for next step
self.concatinate(self.concat_out)
concatinated = update_reads(self.out, 'extendedFrags', 'fastq')
self.input = concatinated
self.assemble_reads(self.assembly_out)
assembled = update_reads(self.assembly_out, 'meta-velvetg', 'fa')
self.step_number += 1
self.input = assembled
return [self.step_number, self.input, concatinated, assembled]