def _make_seq_run():
"""
Make a sequencing run pointed at real data for the tests
:return: SequencingRun object
"""
files_1 = model.SequenceFile([
path.join(path_to_module, "fake_ngs_data", "Data", "Intensities",
"BaseCalls", "01-1111_S1_L001_R1_001.fastq.gz"),
path.join(path_to_module, "fake_ngs_data", "Data", "Intensities",
"BaseCalls", "01-1111_S1_L001_R2_001.fastq.gz"),
])
files_2 = model.SequenceFile([
path.join(path_to_module, "fake_ngs_data", "Data", "Intensities",
"BaseCalls", "02-2222_S1_L001_R1_001.fastq.gz"),
path.join(path_to_module, "fake_ngs_data", "Data", "Intensities",
"BaseCalls", "02-2222_S1_L001_R2_001.fastq.gz"),
])
files_3 = model.SequenceFile([
path.join(path_to_module, "fake_ngs_data", "Data", "Intensities",
"BaseCalls", "03-3333_S1_L001_R1_001.fastq.gz"),
path.join(path_to_module, "fake_ngs_data", "Data", "Intensities",
"BaseCalls", "03-3333_S1_L001_R2_001.fastq.gz"),
])
sample_1 = model.Sample("test_sample", "description", 1)
sample_1.sequence_file = files_1
sample_2 = model.Sample("test_sample", "description", 1)
sample_2.sequence_file = files_2
sample_3 = model.Sample("test_sample", "description", 1)
sample_3.sequence_file = files_3
project = model.Project("test_project", [sample_1, sample_2, sample_3],
"description")
sequencing_run = model.SequencingRun({"layoutType": "PAIRED_END"},
[project], "miseq")
return sequencing_run
def build_sample_list_from_sample_sheet_no_verify(sample_sheet_file):
"""
Create a list of Sample objects, file existence is not verified before SequenceFile is created
this is used when a pre-generated file list is used (e.g. cloud deployment)
:param sample_sheet_file:
:return:
"""
sample_list = _parse_samples(sample_sheet_file)
for sample in sample_list:
sample_dict = sample.get_uploadable_dict()
# create file list
file_list = [sample_dict['File_Forward']]
# if paired end add file to file list
paired_end_read = len(sample_dict['File_Reverse']) > 0
if paired_end_read:
file_list.append(sample_dict['File_Reverse'])
# Create sequence file object and attach to sample
sq = model.SequenceFile(file_list=file_list)
sample.sequence_file = deepcopy(sq)
return sample_list
def parse_sample_list(sample_sheet_file, run_data_directory, run_data_directory_file_list):
"""
Creates a list of Sample Objects
:param sample_sheet_file: Sample Sheet file
:param run_data_directory: Data directory including run directory (e.g. my_run/Data/Intensities/BaseCalls)
:param run_data_directory_file_list: The list of all files in the data directory
:return: list of Sample objects
"""
sample_list = _parse_samples(sample_sheet_file)
for sample in sample_list:
properties_dict = _parse_out_sequence_file(sample)
# this is the Illumina-defined pattern for naming fastq files, from:
# http://blog.basespace.illumina.com/2014/08/18/fastq-upload-in-now-available-in-basespace/
file_pattern = "{sample_name}_S{sample_number}_L\\d{{3}}_R(\\d+)_\\S+\\.fastq.*$".format(
sample_name=re.escape(sample.sample_name), sample_number=sample.sample_number)
logging.info("Looking for files with pattern {}".format(file_pattern))
regex = re.compile(file_pattern)
pf_list = list(filter(regex.search, run_data_directory_file_list))
if not pf_list:
# OK. So we didn't find any files using the **correct** file name
# definition according to Illumina. Let's try again with our deprecated
# behaviour, where we didn't actually care about the sample number:
file_pattern = "{sample_name}_S\\d+_L\\d{{3}}_R(\\d+)_\\S+\\.fastq.*$".format(
sample_name=re.escape(sample.sample_name))
logging.info("Looking for files with pattern {}".format(file_pattern))
regex = re.compile(file_pattern)
pf_list = list(filter(regex.search, run_data_directory_file_list))
if not pf_list:
# we **still** didn't find anything. It's pretty likely, then that
# there aren't any fastq files in the directory that match what
# the sample sheet says...
raise exceptions.SequenceFileError(
("The uploader was unable to find an files with a file name that ends with "
".fastq.gz for the sample in your sample sheet with name {} in the directory {}. "
"This usually happens when the Illumina MiSeq Reporter tool "
"does not generate any FastQ data.").format(
sample.sample_name, run_data_directory))
# List of files may be invalid if directory searching in has been modified by user
if not _validate_pf_list(pf_list):
raise exceptions.SequenceFileError(
("The following file list {} found in the directory {} is invalid. "
"Please verify the folder containing the sequence files matches the SampleSheet file").format(
pf_list, run_data_directory))
# Add the dir to each file to create the full path
for i in range(len(pf_list)):
pf_list[i] = path.join(run_data_directory, pf_list[i])
sq = model.SequenceFile(file_list=pf_list, properties_dict=properties_dict)
sample.sequence_file = deepcopy(sq)
return sample_list
def _parse_sample_list(sample_sheet_file):
"""
Creates a list of all samples in the sample_sheet_file, with accompanying data/metadata
:param sample_sheet_file:
:return: list of samples
"""
sample_list = _parse_samples(sample_sheet_file)
sample_sheet_dir = path.dirname(sample_sheet_file)
base_data_dir = path.join(sample_sheet_dir, "Data", "Intensities",
"BaseCalls")
project_dir_list = next(walk(base_data_dir))[
1] # Get the list of project directories that contain sample files
for sample in sample_list:
project_directory = sample.get('sample_project')
if project_directory not in project_dir_list:
# The project number in the sample sheet does not match a project folder in the run directory
raise exceptions.SequenceFileError(
"The uploader was unable to find the directory '{}' in '{}'. "
"Please verify your SampleSheet Sample_Project matches the directory structure"
.format(project_directory, base_data_dir))
project_data_dir = path.join(base_data_dir, project_directory)
# Create a file list of the data directory, only hit the os once
data_dir_file_list = next(walk(project_data_dir))[2]
properties_dict = _parse_out_sequence_file(sample)
file_pattern = "{sample_name}_S(\\S+)_R(\\d+)_(\\S*)\\.fastq.*$".format(
sample_name=re.escape(sample.sample_name))
logging.info("Looking for files with pattern {}".format(file_pattern))
regex = re.compile(file_pattern)
pf_list = list(filter(regex.search, data_dir_file_list))
if not pf_list:
# we didn't find anything
raise exceptions.SequenceFileError((
"The uploader was unable to find an files with a file name that ends with "
".fastq.gz for the sample in your sample sheet with name {} in the directory {}. "
).format(sample.sample_name, project_data_dir))
# List of files may be invalid if directory searching in has been modified by user
if not _validate_pf_list(pf_list):
raise exceptions.SequenceFileError((
"The following file list {} found in the directory {} is invalid. "
"Please verify the folder containing the sequence files matches the SampleSheet file"
).format(pf_list, project_data_dir))
# Add the dir to each file to create the full path
for i in range(len(pf_list)):
pf_list[i] = path.join(project_data_dir, pf_list[i])
sq = model.SequenceFile(file_list=pf_list,
properties_dict=properties_dict)
sample.sequence_file = deepcopy(sq)
return sample_list
def build_sample_list_from_sample_sheet_with_abs_path(sample_sheet_file):
"""
Create a list of Sample objects, where each SequenceFile object has an absolute file path
:param sample_sheet_file:
:return:
"""
sample_list = _parse_samples(sample_sheet_file)
# Data directory is used if file names on sample sheet are not absolute paths (in directory files)
data_dir = path.dirname(sample_sheet_file)
sample_sheet_dir_file_list = common.get_file_list(data_dir)
for sample in sample_list:
sample_dict = sample.get_uploadable_dict()
paired_end_read = len(sample_dict['File_Reverse']) > 0
# create file list of full paths
file_list = []
# If file is not an abspath already, make it an abspath from filename + data dir
if path.isabs(sample_dict['File_Forward']):
file_list.append(sample_dict['File_Forward'])
elif sample_dict['File_Forward'] in sample_sheet_dir_file_list:
sample_dict['File_Forward'] = path.join(
path.abspath(data_dir), sample_dict['File_Forward'])
file_list.append(sample_dict['File_Forward'])
else:
raise exceptions.SampleSheetError((
"Your sample sheet is malformed. {} Does not match any file in the directory {}"
"".format(sample_dict['File_Forward'], data_dir)),
sample_sheet_file)
# reverse file is same as for forward file
if paired_end_read:
if path.isabs(sample_dict['File_Reverse']):
file_list.append(sample_dict['File_Reverse'])
elif sample_dict['File_Reverse'] in sample_sheet_dir_file_list:
sample_dict['File_Reverse'] = path.join(
path.abspath(data_dir), sample_dict['File_Reverse'])
file_list.append(sample_dict['File_Reverse'])
else:
raise exceptions.SampleSheetError((
"Your sample sheet is malformed. {} Does not match any file in the directory {}"
"".format(sample_dict['File_Reverse'], data_dir)),
sample_sheet_file)
# Create sequence file object and attach to sample
sq = model.SequenceFile(file_list=file_list)
sample.sequence_file = deepcopy(sq)
return sample_list
def test_send_and_get_sequence_files(self):
"""
Tests sending and receiving sequence files
:return:
"""
# upload a project
project_name = "test_project_2"
project_description = "test_project_description"
project = model.Project(name=project_name,
description=project_description)
proj_json_res = self.test_api.send_project(project)
project_identifier = proj_json_res['resource']['identifier']
# upload a sample
sample_name = "test_sample"
sample_desc = "test_sample_desc"
sample = model.Sample(sample_name, sample_desc)
self.test_api.send_sample(sample, project_identifier)
# upload sequence files
sequence_file_list = [
path.join(path_to_module, "fake_dir_data", "file_1.fastq.gz"),
path.join(path_to_module, "fake_dir_data", "file_2.fastq.gz")
]
sequence_file = model.SequenceFile(sequence_file_list)
upload_id = self.test_api.create_seq_run({'layoutType': 'PAIRED_END'},
'miseq')
self.test_api.send_sequence_files(sequence_file, sample_name,
project_identifier, upload_id)
# verify sequence files match what we sent to IRIDA
returned_sequence_files = self.test_api.get_sequence_files(
project_identifier, sample_name)
self.assertEqual(returned_sequence_files[0]['fileName'],
'file_1.fastq.gz')
self.assertEqual(returned_sequence_files[1]['fileName'],
'file_2.fastq.gz')
def parse_sample_list(sample_sheet_file, run_data_directory_file_list):
"""
Creates a list of all sample data in the sample_sheet_file
Verifies data is valid for uploading
:param sample_sheet_file:
:param run_data_directory_file_list: list of all files
:return: list of Sample objects
"""
sample_list = _parse_samples(sample_sheet_file)
data_dir = path.dirname(sample_sheet_file)
data_dir_file_list_full_path = []
for file_name in run_data_directory_file_list:
data_dir_file_list_full_path.append(
path.join(path.abspath(data_dir), file_name))
has_paired_end_read = False
has_single_end_read = False
logging.info(
"Verifying data parsed from sample sheet {}".format(sample_sheet_file))
for sample in sample_list:
sample_dict = sample.get_uploadable_dict()
paired_end_read = len(sample_dict['File_Reverse']) > 0
# keep track if we have both paired and single end reads
if paired_end_read:
has_paired_end_read = True
else:
has_single_end_read = True
# Check if file names are in the files we found in the directory
if ((sample_dict['File_Forward'] not in run_data_directory_file_list)
and
(sample_dict['File_Forward'] not in data_dir_file_list_full_path)):
raise exceptions.SampleSheetError((
"Your sample sheet is malformed. {} Does not match any file in the directory {}"
"".format(sample_dict['File_Forward'], data_dir)),
sample_sheet_file)
if ((paired_end_read and sample_dict['File_Reverse']
not in run_data_directory_file_list)
and (paired_end_read and sample_dict['File_Reverse']
not in data_dir_file_list_full_path)):
raise exceptions.SampleSheetError((
"Your sample sheet is malformed. {} Does not match any file in the directory {}"
"".format(sample_dict['File_Reverse'], data_dir)),
sample_sheet_file)
# create file list of full paths
file_list = []
# Add the dir to each file to create the full path
if sample_dict['File_Forward'] not in data_dir_file_list_full_path:
sample_dict['File_Forward'] = path.join(
data_dir, sample_dict['File_Forward'])
file_list.append(sample_dict['File_Forward'])
if paired_end_read and sample_dict[
'File_Reverse'] not in data_dir_file_list_full_path:
sample_dict['File_Reverse'] = path.join(
data_dir, sample_dict['File_Reverse'])
file_list.append(sample_dict['File_Reverse'])
# Create sequence file object and attach to sample
sq = model.SequenceFile(file_list=file_list)
sample.sequence_file = deepcopy(sq)
# Verify we don't have both single end and paired end reads
if has_single_end_read and has_paired_end_read:
raise exceptions.SampleSheetError(
("Your sample sheet is malformed. "
"SampleSheet cannot have both paired end and single end runs. "
"Make sure all samples are either paired or single."),
sample_sheet_file)
return sample_list
