def merge_split_casava_results(split_dir,output_dir,split_categories=["Index_length","Lane"]):
"""
After split directories are run through casava, this function recombines them. First, the input dirs are
inferred, and then data is moved.
"""
sample_sheet_obj_list = SampleSheetObjList()
sample_sheet_obj_list.__load_sample_sheets_from_meta_directories__(split_dir,split_categories)
#Get all sample ids.
sample_sheet_obj_list = sample_sheet_obj_list.__partition_sample_sheet_objects__("SampleID") #Partition by sample (as casava does)
sample_sheet_obj_list = sample_sheet_obj_list.__partition_sample_sheet_objects__("SampleProject") #Partition by project (as casava does)
#Label th input directory for each sample
for sample_sheet_obj in sample_sheet_obj_list.list:
split_subdir_pieces = []
for category in split_categories:
piece = sample_sheet_obj.__get_meta_datum__(category)
split_subdir_pieces.append(str(piece))
split_subdir = "_".join(split_subdir_pieces)
project_subdir = "Project_" + sample_sheet_obj.__get_meta_datum__("SampleProject")
sample_subdir = "Sample_" + sample_sheet_obj.__get_meta_datum__("SampleID")
original_dir = os.path.join(split_dir,split_subdir+"/"+project_subdir+"/"+sample_subdir)
sample_sheet_obj.__set_meta_datum__("original_dir",original_dir)
#Do the merging
merge_casava_fastq_directories(sample_sheet_obj_list,output_dir,merge_type="move")
move_undetermined_directories_of_min_length(output_dir,sample_sheet_obj_list,"Lane" in split_categories)
return
def merge_flowcell_casava_results(flowcell_dirs,output_dir,*args,**kwargs):
"""
Merges the samples in multiple flowcell directories.
"""
sample_dirs_dict = list_sample_dirs(flowcell_dirs)
sample_sheet_obj_list = SampleSheetObjList()
sample_sheet_obj_list.__load_sample_sheets_from_sample_directories__(sample_dirs_dict)
merge_casava_fastq_directories(sample_sheet_obj_list,output_dir,meta_data_prefix=["FCID"])
return
def __init__(self,config,key=-1,prev_step=None,pipeline=None,split_by_lane=True,split_by_index_length=True,process_name="casava",**kwargs):
"""
In addition to initializing, other steps are completed. These are commented below.
"""
if not prev_step is None:
input_dir = os.path.join(pipeline.output_dir,"Data/Intensities/BaseCalls")
output_dir = os.path.join(pipeline.output_dir,os.path.basename(pipeline.output_dir))
if not os.path.exists(output_dir):
os.makedirs(output_dir)
if pipeline.sample_sheet is None:
original_sample_sheet_file = os.path.join(pipeline.input_dir,"SampleSheet.csv")
else:
original_sample_sheet_file = pipeline.sample_sheet
if not os.path.isfile(original_sample_sheet_file):#Check to make sure original sample sheet exists
send_missing_sample_sheet_email(original_sample_sheet_file)
raise SampleSheetFormatException("No sample sheet found: "+str(original_sample_sheet_file))
sample_sheet_obj_list = SampleSheetObjList(sample_sheet_file=original_sample_sheet_file)
sample_sheet_obj_list.list[0].sample_sheet_table.__write_file__(os.path.join(output_dir,"SampleSheet.csv"))#Copy sample sheet to final output dir.
self.merged = True
split_categories = []
self.split_by_lane = split_by_lane
if split_by_lane is True: #Split by lane (speed up especially for high throughput)
sample_sheet_obj_list = sample_sheet_obj_list.__partition_sample_sheet_objects__("Lane")
split_categories.append("Lane")
self.merged = False
self.split_by_index_length = split_by_index_length
if split_by_index_length == True: #Split by index lane (prevents casava from breaking when pool samples have different index lengths)
for sample_sheet_obj in sample_sheet_obj_list.list:
sample_sheet_obj.__attach_max_column_number__("Index")
sample_sheet_obj_list = sample_sheet_obj_list.__partition_sample_sheet_objects__("Index",use_length=True)
split_categories.append("Index_length")
self.merged = False
number_tasks = len(sample_sheet_obj_list.list)
temporary_output_directories = sample_sheet_obj_list.__create_meta_directories_and_write_files__(os.path.join(output_dir,"split"),split_categories)
self.temporary_output_dir = ":".join(temporary_output_directories)
sample_sheets = [os.path.join(d,"SampleSheet.csv") for d in temporary_output_directories]
self.sample_sheet = ":".join(sample_sheets)
sample_sheet_obj_list.__attach_masks__(run_parameters_path=os.path.join(pipeline.input_dir,"runParameters.xml"))
masks = []
for sample_sheet_obj in sample_sheet_obj_list.list:
mask = sample_sheet_obj.__get_meta_datum__("mask")
mask, number = re.subn(',','-',mask)
masks.append(mask)
self.mask = ":".join(masks)
QsubProcess.__init__(self,config,key=key,output_dir=output_dir,input_dir=input_dir,number_tasks=number_tasks,process_name=process_name,**kwargs)
self.flowcell_key = pipeline.flowcell_key
self.seq_run_key = pipeline.seq_run_key
from processes.hiseq.sample_sheet import SampleSheetObjList
from processes.hiseq.scripts import list_sample_dirs
import argparse
if __name__ == '__main__':
#Handle arguments
parser = argparse.ArgumentParser(description='Test various functions in this functions in this folder that require multiple modules')
parser.add_argument('--load_samples_sample_sheets', dest="samples_dir", type=str, help='Test the loading of sample sheets by sample by providing the path for under which all sub-directories are evaluated for SampleSheet.csv.')
parser.add_argument('--column_values', dest="values_dir", type=str, help='Test the column values function by returning a list of samples in all of the sample sheets by providing the path for under which all sub-directories are evaluated for SampleSheet.csv.')
parser.add_argument('--merge_to_single', dest="merge_dir", type=str, help='Test the merge all sample sheet objects function by returning a single sample sheet by providing the path for under which all sub-directories are evaluated for SampleSheet.csv.')
parser.add_argument('--filter_by_sample', dest="filter_dir", type=str, help='Test the filter sample sheet object by printing multiple sample sheet objects after providing the path for under which all sub-directories are evaluated for SampleSheet.csv.')
args = parser.parse_args()
sample_sheet_obj_list = SampleSheetObjList()
if args.samples_dir:
sample_dirs_dict = list_sample_dirs([args.samples_dir])
sample_sheet_obj_list. __load_sample_sheets_from_sample_directories__(sample_dirs_dict)
sample_sheet_obj_list.__print__()
if args.values_dir:
sample_dirs_dict = list_sample_dirs([args.values_dir])
sample_sheet_obj_list. __load_sample_sheets_from_sample_directories__(sample_dirs_dict)
print str(sample_sheet_obj_list.__get_column_values__("SampleID"))
if args.merge_dir:
sample_dirs_dict = list_sample_dirs([args.merge_dir])
sample_sheet_obj_list. __load_sample_sheets_from_sample_directories__(sample_dirs_dict)
new_sample_sheet_obj_list = sample_sheet_obj_list.__merge_all_sample_sheet_objects__()
new_sample_sheet_obj_list.__print__(print_meta_data=False)
if args.filter_dir:
sample_dirs_dict = list_sample_dirs([args.filter_dir])
sample_sheet_obj_list. __load_sample_sheets_from_sample_directories__(sample_dirs_dict)
sample_ids = sample_sheet_obj_list.__get_column_values__("SampleID")
