def parsingCLIP(args):
"""Parse the CLIP (BG) BAM files."""
GeneAnnotation = gffutils.FeatureDB(args.db_file, keep_order=True)
LoadReads.load_data(
bam_files=args.clip_libs,
genome_dir=args.genome_dir,
gene_annotation=GeneAnnotation,
out_file=args.out_file,
Collapse=args.collapsed,
mask_flank_variants=args.mask_flank_variants,
max_mm=args.max_mm,
ign_out_rds=args.ign_out_rds,
rev_strand=args.rev_strand,
CLIP_exp=True,
min_coverage=int(args.min_coverage),
min_peak=int(args.min_peak)
)
Sequences = LoadReads.get_data_handle(args.out_file, write=True)
if args.mask_miRNA:
print('Removing miRNA-coverage')
ParsingPositions.mask_miRNA_positions(Sequences, GeneAnnotation)
if args.mask_ovrlp:
print('Masking overlapping positions')
ParsingPositions.mask_overlapping_positions(Sequences, GeneAnnotation)
def parsing_files(args, params):
"""Parse arguments that are function of the data files."""
# Loading Sequence and Background
Sequences = LoadReads.get_data_handle(params['dat_file_clip'], write=True)
Background = LoadReads.get_data_handle(params['dat_file_bg'], write=True)
# Estimate the library size
params['LibrarySize'] = tools.estimate_library_size(Sequences)
params['BckLibrarySize'] = tools.estimate_library_size(Background)
# Estimate to number of replicates
gene = list(Sequences.keys())[0]
params['NrOfReplicates'] = len(Sequences[gene]['Coverage'])
params['NrOfBckReplicates'] = len(Background[gene]['Coverage'])
# Initialise the parameter vector alpha
alphashape = (Sequences[gene]['Variants']['0']['shape'][0] +
Sequences[gene]['Coverage']['0'][()].shape[0] +
Sequences[gene]['Read-ends']['0'][()].shape[0])
for state in range(params['NrOfStates']):
params['Diag_event_params']['alpha'][state] = np.random.uniform(
0.9, 1.1, size=(alphashape, args.nr_mix_comp))
return params