def mapper(options, line):
bedtool = pybedtools.BedTool(line, from_string=True)
# loads in the last bedline, because bedtools doesn't have a .next()
for bedline in bedtool:
pass
if options.premRNA:
length = bedline.stop - bedline.start
else:
length = sum(
[int(x) for x in bedline[10][:-1].strip().split(",")]
) # Just gets the lengths of the exons (although no mention of cds or not... not important)
print call_peaks(
[bedline.chrom, bedline.name, bedline.start, bedline.stop, bedline.strand],
length,
None,
options.bam,
int(options.margin),
options.FDR_alpha,
options.threshold,
int(options.minreads),
options.poisson_cutoff,
options.plotit,
10,
1000,
options.SloP,
False,
)
def mapper(options, line):
bedtool = pybedtools.BedTool(line, from_string=True)
#loads in the last bedline, because bedtools doesn't have a .next()
for bedline in bedtool:
pass
if options.premRNA:
length = bedline.stop - bedline.start
else:
length = sum([int(x) for x in bedline[10][:-1].strip().split(",")]) #Just gets the lengths of the exons (although no mention of cds or not... not important)
print call_peaks([bedline.chrom, bedline.name, bedline.start, bedline.stop,
bedline.strand], length, options.bam, int(options.max_gap),
options.FDR_alpha, options.threshold,
int(options.minreads), options.poisson_cutoff,
options.plotit, 10, 1000, options.SloP, False)
def main(options):
##############################################
# logging.info("options : {}".format(options))
##############################################
check_for_index(options.bam)
if options.np == 'autodetect':
options.np = multiprocessing.cpu_count()
pool = multiprocessing.Pool(int(options.np))
bamfile = options.bam
if os.path.exists(bamfile):
#re-set to include the full path to bamfile
bamfile = os.path.abspath(bamfile)
logging.info("bam file is set to %s\n" % (bamfile))
else:
logging.error("Bam file: %s is not defined" % (bamfile))
raise IOError
if options.gtfFile:
# TODO always False - no longer an option
bedtool = build_transcript_data_gtf(
pybedtools.BedTool(options.gtfFile), options.premRNA)
else:
bedtool = build_transcript_data_gtf_as_structure(
options.species, options.premRNA)
bedtool.saveas()
#gets a bedtool of all genes to call peaks on
if options.gene:
bedtool = bedtool.filter(lambda x: x.attrs['gene_id'] in options.gene)
# options.maxgenes # truncates for max bedtool
if options.maxgenes:
########################################################################
logging.info(" number of genes before maxing : {}".format(
len(bedtool)))
logging.info(" max genes from user input: {}".format(options.maxgenes))
########################################################################
if options.maxgenes < len(bedtool):
bedtool = bedtool.random_subset(int(options.maxgenes))
else:
logging.info(
" number of genes <= max genes from user , not truncating genes"
)
pass
exons = get_exon_bed(options.species)
bedtool = bedtool.saveas()
tasks = [
(
bedtool_interval,
bedtool_interval.attrs['effective_length'],
bamfile,
options.max_gap,
options.FDR_alpha,
options.threshold,
options.binom,
options.method,
options.minreads,
options.poisson_cutoff,
options.plotit,
10,
1000,
options.SloP,
options.max_width,
options.min_width,
options.algorithm,
# TODO options.algorithm now always "spline" !
options.reverse_strand,
exons) for gene_no, bedtool_interval in enumerate(bedtool)
]
##################################
# print("len(tasks):", len(tasks))
##################################
#jobs = []
peaks_dicts = []
# generate list of all peaks_dict's, (one peaks_dict per gene)
##############################################################
if options.debug:
peaks_dicts = [call_peaks(*task) for task in tasks]
else:
jobs = [pool.apply_async(call_peaks, task) for task in tasks]
for job, task in zip(jobs, tasks):
try:
peaks_dicts.append(job.get(timeout=options.timeout))
except multiprocessing.TimeoutError as error:
print()
####################################################################################################################################
logging.error(
"gene %s timed out after %s minutes on bedinterval: %s" %
(task[0].attrs['gene_id'], options.timeout / 60, task[0]))
####################################################################################################################################
pool.close()
#################################################################
logging.info("finished call_peaks on all genes")
#################################################################
############################################################
logging.info(" starting adding up transcriptome-wise reads")
############################################################
transcriptome_reads = count_transcriptome_reads(peaks_dicts)
transcriptome_size = count_transcriptome_length(peaks_dicts)
####################################################################################
logging.info(" transcriptome size in bases: {}".format(transcriptome_size))
logging.info(
" transcriptome total number of reads: {}".format(transcriptome_reads))
####################################################################################
filtered_peak_bedtool_tsv = filter_peaks_dicts(peaks_dicts,
options.poisson_cutoff,
transcriptome_size,
transcriptome_reads,
options.use_global_cutoff,
options.bonferroni_correct,
options.algorithm,
options.SloP,
options.min_width,
bypassfiltering=False)
##########################################################
# logging.info(" 1: {}".format(filtered_peak_bedtool_tsv))
##########################################################
###############
# writing files
###############
# options.outfileF, options.save_pickle
#======================================
outbedF = options.outfileF
wether_to_save_pickle = options.save_pickle
#
# writing tsv files
#==================
with open(outbedF + ".tsv", 'w') as tsvfile:
tsvfile.write(filtered_peak_bedtool_tsv)
#
# writing bed files
#==================
pybedtools.BedTool(filtered_peak_bedtool_tsv,
from_string=True).sort(stream=True).saveas(outbedF)
########################################################
#logging.info(" wrote filtered peaks to %s" % (outbedF))
########################################################
#
# writing pickle files
#=====================
if wether_to_save_pickle is True:
with open(
outbedF + ".pickle", 'w'
) as picklefile: # TODO Can't pickle save after filtering ? as we have a tsv now, not a peaks_dicts list !?
pickle.dump(peaks_dicts, file=picklefile)
def func_star(varables):
""" covert f([1,2]) to f(1,2) """
return call_peaks(*varables)
def main(options):
"""
Run the whole pipeline
:rtype: None
"""
##############################################
# logging.info("options : {}".format(options))
##############################################
############ CHECKING FILE STATUS ############
check_for_index(options.bam)
bamfile = options.bam
if os.path.exists(bamfile):
# re-set to include the full path to bamfile
bamfile = os.path.abspath(bamfile)
logging.info("bam file is set to %s\n" % (bamfile))
else:
logging.error("Bam file: %s is not defined" % (bamfile))
raise IOError
########### PREPARE GENE LENGTH ################
# if options.gtfFile:
# # TODO always False - no longer an option
# bedtool = build_transcript_data_gtf(pybedtools.BedTool(options.gtfFile), options.premRNA)
# else:
bedtool = build_transcript_data_gtf_as_structure(options.species, options.premRNA).saveas()
# gets a bedtool of all genes to call peaks on
if options.gene:
bedtool = bedtool.filter(lambda x: x.attrs['gene_id'].split('.')[0] in options.gene).saveas() ### bug
# options.maxgenes # truncates for max bedtool
if options.maxgenes:
logging.info(" number of genes before maxing : {}".format(len(bedtool)))
logging.info(" max genes from user input: {}".format(options.maxgenes))
########################################################################
if options.maxgenes < len(bedtool):
bedtool = bedtool.random_subset(int(options.maxgenes)).saveas()
else:
logging.info(" number of genes <= max genes from user , not truncating genes")
pass
if len(bedtool) == 0:
raise Warning('Bedtool length is 0; check gene id')
exons = get_exon_bed(options.species)
############### PREPARE MULTIPROCESSING ##############################
if options.np == 'autodetect':
options.np = multiprocessing.cpu_count()
pool = multiprocessing.Pool(int(options.np))
tasks = [(bedtool_interval, bedtool_interval.attrs['effective_length'], bamfile, options.max_gap, options.FDR_alpha,
options.threshold, options.binom, options.method, options.minreads, options.poisson_cutoff,
options.plotit, 10, 1000, options.SloP, options.max_width,
options.min_width, options.algorithm,
# TODO options.algorithm now always "spline" !
options.reverse_strand, exons
)
for gene_no, bedtool_interval in enumerate(bedtool)]
logging.info('Total tasks: {}'.format(len(tasks)))
############## CALL PEAKS BY HEIGHT AND CURVE##########################
# generate list of all peaks_dict's, (one peaks_dict per gene)
peaks_dicts = []
if options.debug:
peaks_dicts = [call_peaks(*task) for task in tasks]
else:
jobs = [pool.apply_async(call_peaks, task) for task in tasks]
for job, task in zip(jobs, tasks):
try:
peaks_dicts.append(job.get(timeout=options.timeout))
except multiprocessing.TimeoutError as error:
print()
logging.error("gene %s timed out after %s minutes on bedinterval: %s" % (
task[0].attrs['gene_id'], options.timeout / 60, task[0]))
pool.close()
logging.info("finished call_peaks on all genes")
################### FILTER PEAK BY READ #################################
logging.info(" starting adding up transcriptome-wise reads")
transcriptome_reads = count_transcriptome_reads(peaks_dicts)
transcriptome_size = count_transcriptome_length(peaks_dicts)
logging.info(" transcriptome size in bases: {}".format(transcriptome_size))
logging.info(" transcriptome total number of reads: {}".format(transcriptome_reads))
####################################################################################
filtered_peak_bedtool_tsv = filter_peaks_dicts(peaks_dicts,
options.poisson_cutoff,
transcriptome_size,
transcriptome_reads,
options.use_global_cutoff,
options.bonferroni_correct,
options.algorithm,
options.SloP,
options.min_width,
bypassfiltering=False)
############### WRITE TO FILE #####################################
if type(filtered_peak_bedtool_tsv) == str:
# with open(outbedF + ".tsv", 'w') as tsvfile:
# tsvfile.write(filtered_peak_bedtool_tsv)
# filtered_peak_bedtool_dataframe.to_csv(tsvfile, sep = '\t')
pybedtools.BedTool(filtered_peak_bedtool_tsv, from_string=True).sort(stream=True).saveas(options.outfileF)
if options.save_pickle is True:
with open(options.outfileF + ".pickle", 'w') as f:
# TODO Can't pickle save after filtering ? as we have a tsv now, not a peaks_dicts list !?
pickle.dump(peaks_dicts, file=f)
def func_star(varables):
""" covert f([1,2]) to f(1,2) """
return call_peaks(*varables)
