def grab_flanking_regions_from_mutantfile(
mutant_dataset_infile,
genome,
flanksize=200,
padding_char=".",
min_readcount=0,
chromosome_check_function=lambda x: True,
ignore_both_strand_mutants=False,
):
""" Return (flanking_seq,readcount) with both-side genomic flanking sequences for insertional mutants in mutant_dataset_infile.
Grab all the insertion positions from mutant_dataset_infile (pickled mutant_analysis_classes.Insertional_mutant_dataset object),
use genome (a chrom_name:seq dict) to figure out the flanksize-length flanking sequences on both sides
(padded with padding_char if the end of the chromosome is too close), reverse-complement if needed (if strand=='-')
to get it in the same orientation as the insertion.
Filter the mutants:
- by readcount - ignore mutants with total readcount below min_readcount=0
- by chromosome - ignore mutants in chromosomes for which chromosome_check_function returns False
- by strand - both-strand (merged tandem) mutants will be ignored if ignore_both_strand_mutants is True,
otherwise ValueError will be raised; ValueError will be raised for other unexpected strand values.
For all remaining mutants, append (flanking region seq, total_readcount) to output list.
"""
dataset = mutant_analysis_classes.read_mutant_file(mutant_dataset_infile)
flanking_region_count_list = []
for mutant in sorted(dataset, key=lambda m: m.position):
# filter out mutants with wrong readcounts or in wrong chromosomes
if not chromosome_check_function(mutant.position.chromosome):
continue
if mutant.total_read_count < min_readcount:
continue
# filter out both-stranded mutants if desired;
if mutant.position.strand not in "+-":
if mutant.position.strand == "both" and ignore_both_strand_mutants:
continue
else:
raise ValueError("Unexpected mutant strand! %s" % mutant.position)
# grab mutant position/chromosome
position_before_insertion = mutant.position.min_position
# ignore cassette tandems (i.e. insertions that map to start or end of cassette)
if mutant_analysis_classes.is_cassette_chromosome(mutant.position.chromosome):
if position_before_insertion in [0, len(genome[mutant.position.chromosome])]:
continue
# grab the actual flanking sequence, with padding, correct orientation etc
full_flanking_seq = flanking_region_from_pos(
position_before_insertion,
mutant.position.chromosome,
mutant.position.strand,
genome,
flanksize,
padding_char,
)
# append the sequence and readcount to output data
flanking_region_count_list.append((full_flanking_seq, mutant.total_read_count))
return flanking_region_count_list
def merge_dataset_files(file_list=[], file_glob_pattern=None):
""" Return single mutant dataset from adding all the inpu ones together (input can be filename list or glob pattern).
"""
if (file_list
and file_glob_pattern) or not (file_list or file_glob_pattern):
raise Exception(
"Must provide exactly one of file_list and file_glob_pattern!")
if file_glob_pattern:
file_list = glob.glob(file_glob_pattern)
# make empty dataset
full_dataset = mutant_analysis_classes.Insertional_mutant_pool_dataset()
# read each listed dataset file, merge into the full dataset, then delete single dataset
for infile in file_list:
curr_dataset = mutant_analysis_classes.read_mutant_file(infile)
full_dataset.merge_other_dataset(curr_dataset)
del curr_dataset
return full_dataset
def main(infiles, outfile, options):
""" Run the main functionality of the module (see module docstring for more information), excluding testing.
Print final dataset to outfile (if given); return final multi-dataset object and the list of dataset names in order.
The options argument should be generated by an optparse parser.
"""
# parse all infiles, print summaries to stdout if requested
all_datasets = {}
if options.dataset_names:
dataset_names = options.dataset_names.split(',')
if not len(dataset_names)==len(infiles):
raise ValueError("If dataset names are provided via -D option, you must provide the same number of names "
+"as the total number of infiles! We have %s names and %s infiles."%(len(dataset_names),
len(infiles)))
else:
dataset_names = [os.path.splitext(os.path.basename(infile))[0] for infile in infiles]
for dataset_name,infile in zip(dataset_names,infiles):
if options.verbosity_level>1: print "parsing input file %s - time %s."%(infile, time.ctime())
if infile.endswith('.pickle'):
current_dataset = unpickle(infile)
else:
current_dataset = mutant_analysis_classes.Insertional_mutant_pool_dataset(infile=infile)
current_dataset.count_adjacent_mutants(OUTPUT=None)
# note - read_data_from_file doesn't deal with merging/counting info, so that will be wrong/missing
all_datasets[dataset_name] = current_dataset
if options.verbosity_level>0: print "%s mutants in file %s"%(len(current_dataset), infile)
elif options.verbosity_level>1: current_dataset.print_summary()
# merge datasets into one multi-dataset object
if options.verbosity_level>1: print "merging the mutant data into combined dataset - time %s."%(time.ctime())
multi_dataset = mutant_analysis_classes.Insertional_mutant_pool_dataset(multi_dataset=True)
multi_dataset.populate_multi_dataset(all_datasets, overwrite=False, check_gene_data=True)
# make sure the datasets are in the same order as they were given on the command-line
# (using all_datasets to initialize multi_dataset didn't give an order, since all_datasets is a dictionary)
multi_dataset.dataset_order = dataset_names
# print varying amounts of summary data to stdout
if options.verbosity_level>0: print "total %s mutants present in combined dataset"%(len(multi_dataset))
elif options.verbosity_level>0: multi_dataset.print_summary()
### optionally remove mutants based on another dataset
if options.remove_mutants_from_file:
other_dataset = mutant_analysis_classes.read_mutant_file(options.remove_mutants_from_file)
old_N = len(multi_dataset)
multi_dataset.remove_mutants_based_on_other_dataset(other_dataset,
readcount_min=options.remove_mutants_readcount_min, perfect_reads=options.remove_mutants_min_is_perfect)
if options.verbosity_level>0:
new_N = len(multi_dataset)
print "removed %s mutants based on %s - %s mutants remaining in combined dataset"%(old_N - new_N,
options.remove_mutants_from_file, new_N)
# if requested, add gene annotation info from separate file
if options.gene_annotation_file:
if options.verbosity_level>1:
print "adding gene annotation from file %s - time %s."%(options.gene_annotation_file, time.ctime())
multi_dataset.add_gene_annotation(options.gene_annotation_file,
if_standard_Cre_file=options.annotation_file_is_standard)
# print full data to outfile, unless there is no outfile name given
if outfile:
if options.verbosity_level>1:
print "printing combined dataset output to file %s - time %s."%(outfile, time.ctime())
with open(outfile,'w') as OUTFILE:
write_header_data(OUTFILE,options)
OUTFILE.write("### DATASET SUMMARIES:\n")
multi_dataset.print_summary(OUTPUT=OUTFILE, line_prefix="# ", header_prefix="## ")
OUTFILE.write("### HEADER AND DATA:\n")
multi_dataset.print_data(OUTPUT=OUTFILE, sort_data_by=options.sort_data_key, header_line=True)
# TODO make a *.pickle outfile as well?
return multi_dataset, dataset_names
def main(infiles, outfile, options):
""" Run the main functionality of the module (see module docstring for more information), excluding testing.
The options argument should be generated by an optparse parser.
"""
### parse/process/reformat some options
options.ignore_cassette |= options.separate_cassette
options.count_cassette = not options.dont_count_cassette
options.count_other = not options.dont_count_other
options.merge_boundary_features = not options.dont_merge_boundary_features
# MAYBE-TODO change outfile to a folder? Since it'll have three things in it now...
outfile_basename = os.path.splitext(outfile)[0]
mutant_merging_outfile = outfile_basename + '_merging-info.txt'
# MAYBE-TODO let -C take an optional argument to put the cassette files elsewhere?
cassette_outfile = outfile_basename + '_cassette.txt'
cassette_merging_outfile = outfile_basename + '_cassette_merging-info.txt'
### generate empty alignment set object with basic read position/orientation properties defined by options
if options.Carette:
dataset_class = mutant_Carette.Insertional_mutant_pool_dataset_Carette
else:
dataset_class = mutant_analysis_classes.Insertional_mutant_pool_dataset
all_alignment_data = dataset_class(options.read_cassette_end,
options.read_direction == 'reverse')
if options.separate_cassette:
cassette_alignment_data = dataset_class(
options.read_cassette_end, options.read_direction == 'reverse')
# MAYBE-TODO refactor the whole bunch of "if options.separate_cassette:" clauses to avoid code duplication?
### parse preprocessing/alignment metadata file to get discarded/not-aligned/etc readcounts, pass to all_alignment_data
# (all_alignment_data initializes them to 'unkown', so if file is not given or can't be found/parsed, do nothing)
N_discarded, N_wrong_start, N_no_cassette, N_other_end, N_non_aligned, N_unaligned, N_multiple = \
get_info_from_metadata_files(infiles, options.input_metadata_file, options.read_cassette_end, options.verbosity_level)
if 'unknown' not in (N_wrong_start, N_no_cassette):
assert N_discarded == N_wrong_start+N_no_cassette+N_other_end,\
"Discarded subtotals don't add up to discarded total! %s+%s+%s != %s"%(N_wrong_start,N_no_cassette,
N_other_end,N_discarded)
all_alignment_data.summary.add_discarded_reads(N_discarded, N_wrong_start,
N_no_cassette, N_other_end)
if options.separate_cassette:
cassette_alignment_data.summary.add_discarded_reads(
N_discarded, N_wrong_start, N_no_cassette, N_other_end)
all_alignment_data.summary.add_nonaligned_reads(N_non_aligned, N_unaligned,
N_multiple)
if options.separate_cassette:
cassette_alignment_data.summary.add_nonaligned_reads(
N_non_aligned, N_unaligned, N_multiple)
# MAYBE-TODO also get the final total number of reads from the metadata infile and make sure it's the same
# as the number of processed reads I get from all_alignment_data.print_summary()?
### parse input file and store data - the add_alignment_reader_to_data function here does pretty much all the work!
for infile in infiles:
# if this is a new-style *_genomic-unique.sam file and has a matching *_cassette.sam file, parse that file too
part_infiles = [infile]
if infile.endswith('_genomic-unique.sam'):
cassette_file = infile[:-len('_genomic-unique.sam'
)] + '_cassette.sam'
if os.path.exists(cassette_file):
part_infiles.append(cassette_file)
for part_infile in part_infiles:
# initialize a parser for the SAM infile
if options.verbosity_level > 1:
print "parsing input file %s - time %s." % (part_infile,
time.ctime())
infile_reader = HTSeq.SAM_Reader(part_infile)
# fill the new alignment set object with data from the infile parser
all_alignment_data.add_alignment_reader_to_data(
infile_reader,
uncollapse_read_counts=options.input_collapsed_to_unique,
ignore_cassette=options.ignore_cassette,
cassette_only=False,
treat_unknown_as_match=options.treat_unknown_as_match)
if options.separate_cassette:
cassette_alignment_data.add_alignment_reader_to_data(
infile_reader,
uncollapse_read_counts=options.input_collapsed_to_unique,
ignore_cassette=False,
cassette_only=True,
treat_unknown_as_match=options.treat_unknown_as_match)
### optionally remove mutants based on another dataset - BEFORE adjacent mutant counting/merging
# remove mutants that ARE present in another file (also do it for cassette mutants if those are separate)
if options.remove_mutants_from_file:
for other_file in options.remove_mutants_from_file.split(','):
other_dataset = mutant_analysis_classes.read_mutant_file(
other_file)
all_alignment_data.remove_mutants_in_other_dataset(
other_dataset,
readcount_min=options.remove_from_file_readcount_min,
perfect_reads=options.remove_from_file_min_is_perfect)
if options.separate_cassette:
cassette_alignment_data.remove_mutants_in_other_dataset(
other_dataset,
readcount_min=options.remove_from_file_readcount_min,
perfect_reads=options.remove_from_file_min_is_perfect)
# remove mutants that are NOT present in another file (also do it for cassette mutants if those are separate)
# TODO should I implement using multiple files here too?
if options.remove_mutants_not_from_file:
other_dataset = mutant_analysis_classes.read_mutant_file(
options.remove_mutants_not_from_file)
all_alignment_data.remove_mutants_not_in_other_dataset(
other_dataset,
readcount_min=options.remove_not_from_file_readcount_min,
perfect_reads=options.remove_not_from_file_min_is_perfect)
if options.separate_cassette:
cassette_alignment_data.remove_mutants_not_in_other_dataset(
other_dataset,
readcount_min=options.remove_not_from_file_readcount_min,
perfect_reads=options.remove_not_from_file_min_is_perfect)
### optionally merge some mutant categories
with open(mutant_merging_outfile, 'w') as MERGEFILE:
with open(cassette_merging_outfile, 'w') as CASSETTE_MERGEFILE:
# 1) adjacent same-strand mutants (since they're probably just artifacts of indels during deepseq/PCR)
if options.merge_adjacent_mutants:
try:
leave_N_mutants = int(options.merge_adjacent_leave_mutants)
except ValueError:
leave_N_mutants = options.merge_adjacent_leave_mutants
if options.merge_adjacent_count_ratio is not None:
leave_N_mutants = 'use_ratio'
all_alignment_data.merge_adjacent_mutants(
merge_max_distance=options.adjacent_max_distance,
leave_N_mutants=leave_N_mutants,
min_count_ratio=options.merge_adjacent_count_ratio,
leave_method=options.merge_mutant_choice_method,
merge_cassette_chromosomes=options.merge_in_cassette,
merge_other_chromosomes=options.merge_in_other_chrom,
OUTPUT=MERGEFILE)
if options.merge_in_cassette and options.separate_cassette:
cassette_alignment_data.merge_adjacent_mutants(
merge_max_distance=options.adjacent_max_distance,
leave_N_mutants=leave_N_mutants,
min_count_ratio=options.merge_adjacent_count_ratio,
leave_method=options.merge_mutant_choice_method,
merge_cassette_chromosomes=True,
merge_other_chromosomes=False,
OUTPUT=CASSETTE_MERGEFILE)
# 2) opposite-strand same-position mutants (since they're probably just tail-to-tail cassette tandems)
if options.merge_opposite_tandem_mutants:
try:
leave_N_mutants = int(options.merge_opposite_leave_mutants)
except ValueError:
leave_N_mutants = options.merge_opposite_leave_mutants
if options.merge_opposite_count_ratio is not None:
leave_N_mutants = 'use_ratio'
all_alignment_data.merge_opposite_tandem_mutants(
leave_N_mutants=leave_N_mutants,
max_count_ratio=options.merge_opposite_count_ratio,
leave_method=options.merge_mutant_choice_method,
merge_cassette_chromosomes=options.merge_in_cassette,
merge_other_chromosomes=options.merge_in_other_chrom,
OUTPUT=MERGEFILE)
if options.merge_in_cassette and options.separate_cassette:
cassette_alignment_data.merge_opposite_tandem_mutants(
leave_N_mutants=leave_N_mutants,
max_count_ratio=options.merge_opposite_count_ratio,
leave_method=options.merge_mutant_choice_method,
merge_cassette_chromosomes=True,
merge_other_chromosomes=False,
OUTPUT=CASSETTE_MERGEFILE)
# 3) count adjacent mutants, even if not doing any merging
# Actually there's always a count done after each merge, but that count doesn't print the details to anything,
# so just run it again here with detail-printing - MAYBE-TODO find a more efficient way instead of counting twice?
all_alignment_data.count_adjacent_mutants(
max_distance_to_print=options.adjacent_max_distance,
max_distance_to_count=10000,
count_cassette_chromosomes=options.merge_in_cassette,
count_other_chromosomes=options.merge_in_other_chrom,
OUTPUT=MERGEFILE)
if options.separate_cassette:
cassette_alignment_data.count_adjacent_mutants(
max_distance_to_print=options.adjacent_max_distance,
max_distance_to_count=10,
count_cassette_chromosomes=True,
count_other_chromosomes=False,
OUTPUT=CASSETTE_MERGEFILE)
# MAYBE-TODO make an option for max_distance_to_count? I'm using a much lower one for cassette because it's so dense.
# since there's no optional as/with statement, just remove the cassette_merging_outfile if unwanted
if not options.separate_cassette:
os.remove(cassette_merging_outfile)
### optionally parse gene position/info files and look up the genes for each mutant in the data
if options.gene_position_reference_file is not None:
genefile = options.gene_position_reference_file
if options.verbosity_level > 1:
print "adding genes from file %s to mutant data - time %s." % (
genefile, time.ctime())
all_alignment_data.find_genes_for_mutants(
genefile,
detailed_features=options.detailed_gene_features,
N_run_groups=options.N_detail_run_groups,
verbosity_level=options.verbosity_level)
# if we have gene info, optionally also add annotation
if options.gene_annotation_file:
if options.verbosity_level > 1:
print "adding gene annotation from file %s - time %s." % (
options.gene_annotation_file, time.ctime())
all_alignment_data.add_gene_annotation(
options.gene_annotation_file,
if_standard_Phytozome_file=options.
annotation_file_standard_type,
print_info=(options.verbosity_level >= 2))
### output data to files
save_dataset_files(all_alignment_data, outfile, options.verbosity_level,
True, options.count_cassette, options.count_other,
options.merge_boundary_features, options.sort_data_key,
options.N_sequences_per_group, options)
# TODO write some info about all the other files that go with this one (pickle, merging-info, *cassette*)
if options.separate_cassette:
save_dataset_files(cassette_alignment_data, cassette_outfile, 0, True,
False, False, options.merge_boundary_features,
options.sort_data_key,
options.N_sequences_per_group, options)
def main(infiles, outfile, options):
""" Run the main functionality of the module (see module docstring for more information), excluding testing.
The options argument should be generated by an optparse parser.
"""
### parse/process/reformat some options
options.ignore_cassette |= options.separate_cassette
options.count_cassette = not options.dont_count_cassette
options.count_other = not options.dont_count_other
options.merge_boundary_features = not options.dont_merge_boundary_features
# MAYBE-TODO change outfile to a folder? Since it'll have three things in it now...
outfile_basename = os.path.splitext(outfile)[0]
mutant_merging_outfile = outfile_basename + '_merging-info.txt'
# MAYBE-TODO let -C take an optional argument to put the cassette files elsewhere?
cassette_outfile = outfile_basename + '_cassette.txt'
cassette_merging_outfile = outfile_basename + '_cassette_merging-info.txt'
### generate empty alignment set object with basic read position/orientation properties defined by options
if options.Carette: dataset_class = mutant_Carette.Insertional_mutant_pool_dataset_Carette
else: dataset_class = mutant_analysis_classes.Insertional_mutant_pool_dataset
all_alignment_data = dataset_class(options.read_cassette_end, options.read_direction=='reverse')
if options.separate_cassette:
cassette_alignment_data = dataset_class(options.read_cassette_end, options.read_direction=='reverse')
# MAYBE-TODO refactor the whole bunch of "if options.separate_cassette:" clauses to avoid code duplication?
### parse preprocessing/alignment metadata file to get discarded/not-aligned/etc readcounts, pass to all_alignment_data
# (all_alignment_data initializes them to 'unkown', so if file is not given or can't be found/parsed, do nothing)
N_discarded, N_wrong_start, N_no_cassette, N_other_end, N_non_aligned, N_unaligned, N_multiple = \
get_info_from_metadata_files(infiles, options.input_metadata_file, options.read_cassette_end, options.verbosity_level)
if 'unknown' not in (N_wrong_start, N_no_cassette):
assert N_discarded == N_wrong_start+N_no_cassette+N_other_end,\
"Discarded subtotals don't add up to discarded total! %s+%s+%s != %s"%(N_wrong_start,N_no_cassette,
N_other_end,N_discarded)
all_alignment_data.summary.add_discarded_reads(N_discarded, N_wrong_start, N_no_cassette, N_other_end)
if options.separate_cassette:
cassette_alignment_data.summary.add_discarded_reads(N_discarded, N_wrong_start, N_no_cassette, N_other_end)
all_alignment_data.summary.add_nonaligned_reads(N_non_aligned, N_unaligned, N_multiple)
if options.separate_cassette:
cassette_alignment_data.summary.add_nonaligned_reads(N_non_aligned, N_unaligned, N_multiple)
# MAYBE-TODO also get the final total number of reads from the metadata infile and make sure it's the same
# as the number of processed reads I get from all_alignment_data.print_summary()?
### parse input file and store data - the add_alignment_reader_to_data function here does pretty much all the work!
for infile in infiles:
# if this is a new-style *_genomic-unique.sam file and has a matching *_cassette.sam file, parse that file too
part_infiles = [infile]
if infile.endswith('_genomic-unique.sam'):
cassette_file = infile[:-len('_genomic-unique.sam')] + '_cassette.sam'
if os.path.exists(cassette_file):
part_infiles.append(cassette_file)
for part_infile in part_infiles:
# initialize a parser for the SAM infile
if options.verbosity_level>1:
print "parsing input file %s - time %s."%(part_infile, time.ctime())
infile_reader = HTSeq.SAM_Reader(part_infile)
# fill the new alignment set object with data from the infile parser
all_alignment_data.add_alignment_reader_to_data(infile_reader,
uncollapse_read_counts = options.input_collapsed_to_unique,
ignore_cassette = options.ignore_cassette, cassette_only = False,
treat_unknown_as_match = options.treat_unknown_as_match)
if options.separate_cassette:
cassette_alignment_data.add_alignment_reader_to_data(infile_reader,
uncollapse_read_counts = options.input_collapsed_to_unique,
ignore_cassette = False, cassette_only = True,
treat_unknown_as_match = options.treat_unknown_as_match)
### optionally remove mutants based on another dataset - BEFORE adjacent mutant counting/merging
# remove mutants that ARE present in another file (also do it for cassette mutants if those are separate)
if options.remove_mutants_from_file:
for other_file in options.remove_mutants_from_file.split(','):
other_dataset = mutant_analysis_classes.read_mutant_file(other_file)
all_alignment_data.remove_mutants_in_other_dataset(other_dataset,
readcount_min=options.remove_from_file_readcount_min, perfect_reads=options.remove_from_file_min_is_perfect)
if options.separate_cassette:
cassette_alignment_data.remove_mutants_in_other_dataset(other_dataset,
readcount_min=options.remove_from_file_readcount_min, perfect_reads=options.remove_from_file_min_is_perfect)
# remove mutants that are NOT present in another file (also do it for cassette mutants if those are separate)
# TODO should I implement using multiple files here too?
if options.remove_mutants_not_from_file:
other_dataset = mutant_analysis_classes.read_mutant_file(options.remove_mutants_not_from_file)
all_alignment_data.remove_mutants_not_in_other_dataset(other_dataset,
readcount_min=options.remove_not_from_file_readcount_min, perfect_reads=options.remove_not_from_file_min_is_perfect)
if options.separate_cassette:
cassette_alignment_data.remove_mutants_not_in_other_dataset(other_dataset,
readcount_min=options.remove_not_from_file_readcount_min, perfect_reads=options.remove_not_from_file_min_is_perfect)
### optionally merge some mutant categories
with open(mutant_merging_outfile, 'w') as MERGEFILE:
with open(cassette_merging_outfile, 'w') as CASSETTE_MERGEFILE:
# 1) adjacent same-strand mutants (since they're probably just artifacts of indels during deepseq/PCR)
if options.merge_adjacent_mutants:
try: leave_N_mutants = int(options.merge_adjacent_leave_mutants)
except ValueError: leave_N_mutants = options.merge_adjacent_leave_mutants
if options.merge_adjacent_count_ratio is not None: leave_N_mutants = 'use_ratio'
all_alignment_data.merge_adjacent_mutants(merge_max_distance=options.adjacent_max_distance,
leave_N_mutants=leave_N_mutants, min_count_ratio = options.merge_adjacent_count_ratio,
leave_method=options.merge_mutant_choice_method, merge_cassette_chromosomes = options.merge_in_cassette,
merge_other_chromosomes = options.merge_in_other_chrom, OUTPUT = MERGEFILE)
if options.merge_in_cassette and options.separate_cassette:
cassette_alignment_data.merge_adjacent_mutants(merge_max_distance = options.adjacent_max_distance,
leave_N_mutants=leave_N_mutants, min_count_ratio = options.merge_adjacent_count_ratio,
leave_method=options.merge_mutant_choice_method, merge_cassette_chromosomes = True,
merge_other_chromosomes = False, OUTPUT = CASSETTE_MERGEFILE)
# 2) opposite-strand same-position mutants (since they're probably just tail-to-tail cassette tandems)
if options.merge_opposite_tandem_mutants:
try: leave_N_mutants = int(options.merge_opposite_leave_mutants)
except ValueError: leave_N_mutants = options.merge_opposite_leave_mutants
if options.merge_opposite_count_ratio is not None: leave_N_mutants = 'use_ratio'
all_alignment_data.merge_opposite_tandem_mutants(leave_N_mutants=leave_N_mutants,
max_count_ratio = options.merge_opposite_count_ratio, leave_method=options.merge_mutant_choice_method,
merge_cassette_chromosomes = options.merge_in_cassette,
merge_other_chromosomes = options.merge_in_other_chrom, OUTPUT = MERGEFILE)
if options.merge_in_cassette and options.separate_cassette:
cassette_alignment_data.merge_opposite_tandem_mutants(leave_N_mutants=leave_N_mutants,
max_count_ratio = options.merge_opposite_count_ratio, leave_method=options.merge_mutant_choice_method,
merge_cassette_chromosomes = True, merge_other_chromosomes = False, OUTPUT = CASSETTE_MERGEFILE)
# 3) count adjacent mutants, even if not doing any merging
# Actually there's always a count done after each merge, but that count doesn't print the details to anything,
# so just run it again here with detail-printing - MAYBE-TODO find a more efficient way instead of counting twice?
all_alignment_data.count_adjacent_mutants(max_distance_to_print = options.adjacent_max_distance,
max_distance_to_count = 10000, count_cassette_chromosomes = options.merge_in_cassette,
count_other_chromosomes = options.merge_in_other_chrom, OUTPUT = MERGEFILE)
if options.separate_cassette:
cassette_alignment_data.count_adjacent_mutants(max_distance_to_print = options.adjacent_max_distance,
max_distance_to_count = 10, count_cassette_chromosomes = True,
count_other_chromosomes = False, OUTPUT = CASSETTE_MERGEFILE)
# MAYBE-TODO make an option for max_distance_to_count? I'm using a much lower one for cassette because it's so dense.
# since there's no optional as/with statement, just remove the cassette_merging_outfile if unwanted
if not options.separate_cassette:
os.remove(cassette_merging_outfile)
### optionally parse gene position/info files and look up the genes for each mutant in the data
if options.gene_position_reference_file is not None:
genefile = options.gene_position_reference_file
if options.verbosity_level>1: print "adding genes from file %s to mutant data - time %s."%(genefile, time.ctime())
all_alignment_data.find_genes_for_mutants(genefile, detailed_features=options.detailed_gene_features,
N_run_groups=options.N_detail_run_groups,
verbosity_level=options.verbosity_level)
# if we have gene info, optionally also add annotation
if options.gene_annotation_file:
if options.verbosity_level>1:
print "adding gene annotation from file %s - time %s."%(options.gene_annotation_file, time.ctime())
all_alignment_data.add_gene_annotation(options.gene_annotation_file,
if_standard_Phytozome_file=options.annotation_file_standard_type, print_info=(options.verbosity_level >= 2))
### output data to files
save_dataset_files(all_alignment_data, outfile, options.verbosity_level, True, options.count_cassette, options.count_other,
options.merge_boundary_features, options.sort_data_key, options.N_sequences_per_group, options)
# TODO write some info about all the other files that go with this one (pickle, merging-info, *cassette*)
if options.separate_cassette:
save_dataset_files(cassette_alignment_data, cassette_outfile, 0, True, False, False,
options.merge_boundary_features, options.sort_data_key, options.N_sequences_per_group, options)
