def run():
scriptPath = os.path.dirname(os.path.realpath(__file__))
plot_script = os.path.join(scriptPath, 'plotNucleotideProbabilities.R')
parser = create_parser()
args = parser.parse_args()
functions.checkExistence(args.inputfile)
functions.checkPath(args.outdir)
outfile = os.path.join(args.outdir, args.prefix + '_nuc_mutations.table')
outfile_img = os.path.join(args.outdir, args.prefix + '_nuc_mutations.pdf')
main(args.inputfile, outfile, args.coverage, args.verbose)
cmd = [
'R',
'-q',
'--slave',
'-f %r' % plot_script,
'--args',
'%r' % outfile,
'%r' % outfile_img,
args.limit,
]
execute(cmd)
if args.remove:
os.remove(outfile)
def run():
scriptPath = os.path.dirname(os.path.realpath(__file__))
plot_script = os.path.join(scriptPath, 'plotHeatMapSmall.R')
parser = create_parser()
args = parser.parse_args()
functions.checkExistence(args.parclip)
functions.checkExistence(args.gff)
prefix_pattern = '%s_mat_sm_up%s_do%s_min%s_max%s_xbins%s_ybins%s'
out_prefix = prefix_pattern % (args.prefix, args.upstream, args.downstream,
args.min, args.max, args.xbins, args.ybins)
outfile_mat_sense = os.path.join(args.outputdir,
out_prefix + '_sense.table')
outfile_mat_asense = os.path.join(args.outputdir,
out_prefix + '_asense.table')
outfile_img_sense = os.path.join(args.outputdir, out_prefix + '_sense.png')
outfile_img_asense = os.path.join(args.outputdir,
out_prefix + '_asense.png')
sense = main(args.parclip, args.gff, args.upstream, args.downstream, True,
args.min, args.max, args.verbose, args.xbins, args.ybins,
'Collecting data from sense strand')
asense = main(args.parclip, args.gff, args.upstream, args.downstream,
False, args.min, args.max, args.verbose, args.xbins,
args.ybins, 'Collecting data from anti-sense strand')
total = args.upstream + args.max + 1 + args.downstream
saveMat(outfile_mat_sense, sense[0], args.upstream, args.downstream,
sense[1], args.xbins, total)
saveMat(outfile_mat_asense, asense[0], args.upstream, args.downstream,
asense[1], args.xbins, total)
start = args.upstream / total * args.xbins
cmd = [
'R',
'-q',
'--slave',
'-f %r' % plot_script,
'--args',
'%r' % outfile_mat_sense,
'%r' % outfile_mat_asense,
'%r' % outfile_img_sense,
'%r' % outfile_img_asense,
0.98, # hard-coded qvalue
start,
args.ypx,
args.xpx,
]
execute(cmd)
if args.remove:
os.remove(outfile_mat_sense)
os.remove(outfile_mat_asense)
def cleanup(self, keep_intermed=False):
"""Cleans up temporary and intermediate files
Args:
keep_intermed (:obj:`boolean`): if set to True, intermediate files are not removed
"""
if not self._keep_all:
for rm_file in self._tmp_files:
logger.debug('%s is marked as a temporary file, cleaning up',
rm_file)
execute('rm -rf %s' % rm_file, exit=False)
if not keep_intermed:
for rm_file in self._intermed_files:
logger.debug(
'%s is marked as an intermediate file, cleaning up',
rm_file)
execute('rm -rf %s' % rm_file, exit=False)
def run():
scriptPath = os.path.dirname(os.path.realpath(__file__))
plot_script = os.path.join(scriptPath, 'plotKmerPerPosition.R')
parser = create_parser()
args = parser.parse_args()
sites = ParclipSiteContainer.from_file(args.inputfile)
if args.filterGFF != '':
sites.remove_gff_sites(args.filterGFF, args.awidth)
sites.sort(by=args.key, ascending=False)
with EfficientGenome(args.genome) as genome:
sites = sites[args.start:args.stop]
seqs = sites.get_all_sequences(genome, args.width)
prefix_fmt = '%s_kmerPerPosition_kmer%s_start%s_stop%s_width%s_sort_%s'
prefix = prefix_fmt % (args.prefix, args.kmer, args.start, args.stop,
args.width, args.key)
outfile_table = os.path.join(args.outdir, prefix + '.table')
outfile_pdf = os.path.join(args.outdir, prefix + '.pdf')
seq_len = 2 * args.width + 1
getKmerOccurences(seqs,
seq_len,
outfile_table,
kmer=(args.kmer - 1),
verbose=args.verbose)
cmd = [
'R', '-q', '--slave',
'-f %s' % plot_script, '--args', outfile_table, outfile_pdf,
args.width, 0, args.width + 1
]
execute(cmd)
if args.remove:
os.remove(outfile_table)
def run():
scriptPath = os.path.dirname(os.path.realpath(__file__))
parser = create_parser()
args = parser.parse_args()
prefix_pat = '%s_xxmotif_start%s_stop%s_width%s_sort_%s'
file_prefix = prefix_pat % (args.prefix, args.start, args.stop, args.width,
args.key)
sites = ParclipSiteContainer.from_file(args.inputfile)
if args.filterGFF != '':
sites.remove_gff_sites(args.filterGFF, args.awidth)
sites.sort(by=args.key, ascending=False)
sites = sites[args.start:args.stop]
gen_file = os.path.join(args.outdir, file_prefix + '.fa')
with EfficientGenome(args.genome) as genome:
sites.save2Fasta(genome, gen_file, width=args.width)
cmd = [
'XXmotif',
args.outdir,
gen_file,
'--zoops',
'--merge-motif-threshold LOW',
'--max-match-positions 10',
]
if args.negSet:
cmd.append('--negSet %s' % args.negSet)
execute(cmd)
tmp_dir = os.path.join(args.outdir, 'tmp')
mini_plot_script = os.path.join(tmp_dir, 'plotDistribution.R')
mini_plot_cmd = [
'R',
'-q',
'--slave',
'-f %r' % mini_plot_script,
'--args',
'%r' % args.outdir,
]
execute(mini_plot_cmd)
plot_script = os.path.join(scriptPath, '..', 'plots', 'weblogo.R')
pwm_file = os.path.join(args.outdir, file_prefix + '.pwm')
plot_cmd = [
'R',
'-q',
'--slave',
'-f %s' % plot_script,
'--args',
pwm_file,
args.outdir,
file_prefix,
args.plotPWM,
]
if args.plotPWM > 0:
execute(plot_cmd)
if not args.keep_tmp_files:
shutil.rmtree(tmp_dir, ignore_errors=True)
def execute(self):
"""Execute all queued commands"""
for cmd in self._cmds:
yield execute(cmd)
def main(parclip, outdir, prefix, genomepath, negset, gfffile, kmer, key,
useQuantiles, verbose, args):
scriptPath = os.path.dirname(os.path.realpath(__file__))
plot_script = os.path.join(scriptPath, 'plotKmerLogOdds.R')
pc = ParclipSiteContainer.from_file(parclip)
if gfffile is not None:
pc.remove_gff_sites(gfffile)
pc.sort(by=key, ascending=False)
kmers = functions.makekmers(kmer, list('ACGT'))[kmer - 1]
negfreq = loadNegTable(negset)
with EfficientGenome(genomepath) as genomeseq:
allfreqs = []
fileprefix = '%s_logodds_%smer_sort_%s' % (prefix, kmer, key)
if useQuantiles:
fileprefix = fileprefix + '_quantiles'
allfreqs.append(
getkmerLogs(pc, genomeseq, negfreq, kmers, 0, 1000, 15))
quantiles = [
0.01, 0.02, 0.03, 0.04, 0.05, 0.1, 0.125, 0.15, 0.175, 0.2,
0.225, 0.25, 0.275, 0.3, 0.325, 0.35, 0.375, 0.4, 0.45, 0.5,
0.55, 0.65, 0.7, 0.75, 0.8, 0.85, 0.9
]
count = 1
stop = 1000
for q in quantiles:
if verbose:
functions.showProgress(
count, len(quantiles),
'Getting kmer log-odds from quantiles...')
old_stop = stop
start = functions.getQuantileIndex(len(pc), q) - 500
stop = functions.getQuantileIndex(len(pc), q) + 500
if start < 0:
start = 0
if stop > len(pc) - 2:
break
count = count + 1
if (stop - 500) < old_stop:
msg_pat = 'Bin %s and %s are overlapping by %s sites!'
# TODO 2x quantiles[count - 2] is probably a bug
msg = msg_pat % (quantiles[count - 2],
quantiles[count - 2], old_stop -
(stop - 500))
print(msg, file=sys.stderr)
allfreqs.append(
getkmerLogs(pc, genomeseq, negfreq, kmers, start, stop,
15))
else:
maxsize = 50000
stepsize = 1000
start = 0
stop = 1000
run = True
while run:
if stop > len(pc) - 2 or stop > maxsize:
print()
print('STOP at: %s' % +stop)
run = False
break
if verbose:
functions.showProgress(
stop, maxsize, 'Getting kmer log-odds from bins...')
allfreqs.append(
getkmerLogs(pc, genomeseq, negfreq, kmers, start, stop,
15))
start = stop
stop = stop + stepsize
table_file = os.path.join(outdir, fileprefix + '.table')
pdf_file = os.path.join(outdir, fileprefix + '.pdf')
sortAndSave(allfreqs, table_file, kmers)
cmd = [
'R',
'-q',
'--slave',
'-f %r' % plot_script,
'--args',
'%r' % table_file,
'%r' % pdf_file,
]
execute(cmd)
if not args.keep_tmp_files:
os.remove(table_file)
def main():
parser = create_parser()
args = parser.parse_args()
np.random.seed(args.seed)
pc = ParclipSiteContainer.from_file(args.pc_table)
gff_records = []
with open(args.gff_file) as gff_handle:
parser = GFF3Parser(gff_handle)
for rec in parser.parse():
if args.min_ts_len <= rec.end - rec.start + 1 <= args.max_ts_len:
gff_records.append(rec)
cut_len = args.downstream_bp + args.upstream_bp + 2 * args.gene_bp + 2
gene_bp = args.gene_bp
upstream_bp = args.upstream_bp
downstream_bp = args.downstream_bp
def aggregate_data(gff_rec, sense=True):
def rev(strand):
return '+' if strand == '-' else '-'
chrom = gff_rec.seqid
start = gff_rec.start
end = gff_rec.end
smooth_window = args.smooth_window
anno_strand = gff_rec.strand
query_strand = anno_strand if sense else rev(anno_strand)
if anno_strand == '+':
values_upstream = pc.get_occ_profile(chrom, start - upstream_bp,
start + gene_bp, query_strand)
values_dostream = pc.get_occ_profile(chrom, end - gene_bp,
end + downstream_bp,
query_strand)
else:
values_upstream = pc.get_occ_profile(chrom, end - gene_bp,
end + upstream_bp,
query_strand)[::-1]
values_dostream = pc.get_occ_profile(chrom, start - downstream_bp,
start + gene_bp,
query_strand)[::-1]
upstr = pd.Series(values_upstream).rolling(window=smooth_window,
center=True,
min_periods=0).mean()
dostr = pd.Series(values_dostream).rolling(window=smooth_window,
center=True,
min_periods=0).mean()
return np.hstack((upstr, dostr))
def write_out_data(sense, out_file, bs_file):
global data_dict
data_dict = {}
for i, gff_rec in enumerate(gff_records):
data = aggregate_data(gff_rec, sense)
data_dict[i] = data
ids = np.array(list(data_dict.keys()))
# actual smoothed curve
with open(out_file, 'w') as out:
avg_vec = np.zeros(cut_len)
for ts_ind in ids:
avg_vec += data_dict[ts_ind]
avg_vec /= len(ids)
print(*avg_vec, sep='\t', file=out)
with open(bs_file, 'w') as out:
with Pool(args.n_processes) as pool:
jobs = []
for bs in range(args.n_bs_iterations):
ts_ind = np.random.choice(ids, size=len(ids))
job = pool.apply_async(calc_profile,
args=(ts_ind, cut_len))
jobs.append(job)
for job in jobs:
res_vec = job.get()
print(*res_vec, sep='\t', file=out)
prefix_fmt = '%s_centerBoth_up%s_gene%s_do%s_min%s_max%s'
fmt_args = (
args.prefix,
args.upstream_bp,
args.gene_bp,
args.downstream_bp,
args.min_ts_len,
args.max_ts_len,
)
fn_prefix = prefix_fmt % fmt_args
sense_table = os.path.join(args.outputdir, fn_prefix + '_sense.table')
sense_bs_table = os.path.join(args.outputdir,
fn_prefix + '_sense_bs.table')
asense_table = os.path.join(args.outputdir, fn_prefix + '_asense.table')
asense_bs_table = os.path.join(args.outputdir,
fn_prefix + '_asense_bs.table')
write_out_data(True, sense_table, sense_bs_table)
write_out_data(False, asense_table, asense_bs_table)
scriptPath = os.path.dirname(os.path.realpath(__file__))
plot_script = os.path.join(scriptPath, 'plotCenterBothEnds_bs.R')
if not args.title:
title = args.prefix
else:
title = args.title
outfile_pdf = os.path.join(args.outputdir,
fn_prefix + '_sm%s.pdf' % args.smooth_window)
plot_cmd = [
'R',
'-q',
'--slave',
'-f %r' % plot_script,
'--args',
'%r' % sense_table,
'%r' % sense_bs_table,
'%r' % asense_table,
'%r' % asense_bs_table,
'%r' % outfile_pdf,
'%r' % title,
args.upstream_bp,
args.downstream_bp,
args.gene_bp,
args.smooth_window,
'%r' % args.labelCenterA,
'%r' % args.labelBody,
'%r' % args.labelCenterB,
]
execute(plot_cmd)
if args.cleanup:
os.remove(sense_table)
os.remove(sense_bs_table)
os.remove(asense_table)
os.remove(asense_bs_table)
def run():
scriptPath = os.path.dirname(os.path.realpath(__file__))
plot_script = os.path.join(scriptPath, 'plotCenterBothEnds.R')
parser = create_parser()
args = parser.parse_args()
if not args.title:
title = args.prefix
else:
title = args.title
prefix_fmt = '%s_centerBoth_up%s_gene%s_do%s_min%s_max%s'
fmt_args = (
args.prefix,
args.upstream,
args.gene,
args.downstream,
args.min,
args.max,
)
fn_prefix = prefix_fmt % fmt_args
outfile_sense = os.path.join(args.outputdir, fn_prefix + '_sense.table')
outfile_asense = os.path.join(args.outputdir, fn_prefix + '_asense.table')
outfile_pdf = os.path.join(args.outputdir, fn_prefix + '_sm%s.pdf' % args.plotSmooth)
start_time = datetime.datetime.now()
main(args.parclip, outfile_sense, args.gff, args.downstream, args.upstream,
args.gene, True, args.min, args.max, args.verbose,
'Collecting PAR-CLIP sense data')
main(args.parclip, outfile_asense, args.gff, args.downstream, args.upstream,
args.gene, False, args.min, args.max, args.verbose,
'Collecting PAR-CLIP anti-sense data')
if args.verbose:
end_time = datetime.datetime.now()
run_time = end_time - start_time
print()
print('time: %s seconds' % run_time.seconds)
plot_cmd = [
'R',
'-q',
'--slave',
'-f %r' % plot_script,
'--args',
'%r' % outfile_sense,
'%r' % outfile_asense,
'%r' % outfile_pdf,
'%r' % title,
args.upstream,
args.downstream,
args.gene,
args.plotSmooth,
'%r' % args.labelCenterA,
'%r' % args.labelBody,
'%r' % args.labelCenterB,
]
execute(plot_cmd)
if args.remove:
os.remove(outfile_sense)
os.remove(outfile_asense)