def main():
p = argparse.ArgumentParser(description=__doc__,
formatter_class=argparse.RawDescriptionHelpFormatter)
p.add_argument("-p", dest="pvals", help="BED containing all the p values"
" used to generate `regions`")
p.add_argument("-r", dest="regions", help="BED containing all the regions")
p.add_argument("-s", "--step", dest="step", type=int, default=50,
help="step size for acf calculation. should be the same "
" value as the step sent to -d arg for acf")
p.add_argument("-c", dest="c", help="column number containing the p-value"
" of interest", type=str, default=-1)
p.add_argument("-z", dest="z", help="use z-score correction",
action="store_true")
args = p.parse_args()
if not (args.regions and args.pvals):
import sys
sys.exit(not p.print_help())
header = ts.nopen(args.regions).next()
if header.startswith("#") or (not header.split("\t")[2].isdigit()):
print "%s\tslk_p\tslk_sidak_p" % (header.rstrip("\r\n"),)
header = ts.header(args.pvals)
if args.c in header:
args.c = header.index(args.c) + 1
else:
args.c = int(args.c)
return run(args)
def main(tests, xref, head):
df = pandas.read_table(xref, index_col=[0], compression=get_comp(xref))
a, b = get_ab(tests)
out_order = header(tests)
if head:
print "\t".join(out_order)
for l in reader(tests):
try:
if df[a][l['SiteA']] or df[b][l['SiteA']] or df[a][l['SiteB']] or df[b][l['SiteB']]:
print "\t".join(l[h] for h in out_order)
except KeyError:
# definitely not a primary site
pass
def main(snpeff, descriptions):
snpeff_header = header(snpeff)
additional_columns = header(descriptions)
full_header = snpeff_header + additional_columns
descriptions_dict = descriptions_to_dict(descriptions)
print "\t".join(full_header)
for l in reader(snpeff):
#Gene_ID Gene_name
# l['Gene_ID']
add = dict()
try:
add = descriptions_dict[l['Gene_ID']]
except KeyError:
try:
add = descriptions_dict[l['Gene_name']]
except KeyError:
print "\t".join(l[h] for h in snpeff_header)
continue
full_line = dict(l.items() + add.items())
try:
assert full_line.get('status')
except AssertionError:
full_line['status'] = ""
print "\t".join(full_line[h] for h in full_header)
def uniprot(args):
"""Add Uniprot annotation to gene list. Args: genes, uniprotdb, column"""
uniprot_db = {}
uniprot_header = header(args.uniprotdb)
for entry in reader(args.uniprotdb):
for gene in entry['Gene names'].split():
uniprot_db[gene] = entry
for entry in reader(args.genes, header=False):
uniprot_fields = []
for gene in entry[int(args.column) - 1].split(","):
uniprot = uniprot_db.get(gene)
if uniprot:
for h in uniprot_header:
uniprot_fields.append(uniprot[h])
print "\t".join(entry) + "\t".join(map(str, uniprot_fields))
def get_col_num(c, bed_file=None):
"""
adjust the col number so it does intutive stuff
for command-line interface
>>> get_col_num(4)
3
>>> get_col_num(-1)
-1
"""
if isinstance(c, basestring) and c.isdigit():
c = int(c)
if isinstance(c, (int, long)):
return c if c < 0 else (c - 1)
header = ts.header(bed_file)
assert c in header
return header.index(c)
def main(result_files):
common_header = header(result_files[0])[:-1]
samples = set()
for file in result_files:
samples.add(op.basename(file).rsplit("_", 2)[0])
# leaves one entry per unique sample
for sample in samples:
# open a new file to facilitate the join
out = gzip.open("{sample}.combined.txt.gz".format(sample=sample), "wb")
i = 1
print >>out, "\t".join(common_header)
for result_file in result_files:
if not op.basename(result_file).startswith(sample): continue
for toks in reader(result_file, header=common_header):
if not toks['Functionality'] == 'productive': continue
toks['Sequence number'] = str(i)
print >>out, "\t".join(toks[h] for h in common_header)
i += 1
out.close()
def main(args):
annotation_dict = make_dict(args.annotated_peaks)
gtf = gtf2dict(args.gtf)
blast_header = header(args.blast)
# only save append these columns onto annotation
annotation_save = "baseMeanA baseMeanB pval padj sequence".split()
blast_header.extend(annotation_save)
# stuff one may want from gtf
gtf_save = "symbol ensg enst".split()
blast_header.extend(gtf_save)
print "\t".join(blast_header)
for b in reader(args.blast, header=True):
annotation_lookup = annotation_dict[b['qseqid']]
# ENST00000576218_-
enst = b['sseqid'].split("_")[0]
try:
gtf_lookup = gtf[enst]
except KeyError:
gtf_lookup = {'enst':enst, 'ensg':'-', 'symbol':'-'}
d = dict(b.items() + annotation_lookup.items() + gtf_lookup.items())
print "\t".join([d[i] for i in blast_header])
def main(deseq_output, ensembl_gtf):
p = "%s.pickle" % ensembl_gtf
if os.path.exists(p):
with open(p, 'rb') as handle:
ensembl_translation = pickle.load(handle)
else:
ensembl_translation = make_ens_table(ensembl_gtf)
with open(p, 'wb') as handle:
pickle.dump(ensembl_translation, handle)
h = header(deseq_output, sep=",")
h[0] = 'ensg'
h = [i.strip('"') for i in h]
h.append('type')
print ",".join(h)
for toks in reader(deseq_output, sep=","):
toks['ensg'] = toks['']
toks['type'] = ensembl_translation[toks['ensembl']]
print ",".join(toks[i] for i in h)
def filter(p_bed, region_bed, max_p=None, region_p=None, p_col_name="P.Value", coef_col_name="logFC"):
ph = ts.header(p_bed)
if (ph[1] + ph[2]).isdigit():
raise Exception("need header in p-value file to run filter")
assert ph[1] == "start" and ph[2] == "end" and ph[0] == "chrom", ("must have chrom, start, end header for", p_bed)
ph = ["p" + h for h in ph]
rh = ts.header(region_bed)
header = not (rh[1] + rh[2]).isdigit()
if isinstance(p_col_name, str) and p_col_name.isdigit():
p_col_name = int(p_col_name) - 1
if isinstance(p_col_name, (int, long)):
p_col_name = ph[p_col_name][1:]
a = dict(p_bed=p_bed, region_bed=region_bed)
a["p_bed"] = fix_bed(a["p_bed"])
a["header"] = ""
j = 0
for group, plist in groupby(
ts.reader(
"|bedtools intersect -b %(p_bed)s \
-a %(region_bed)s -wo %(header)s"
% a,
header=rh + ph,
),
itemgetter("chrom", "start", "end"),
):
plist = list(plist)
if region_p:
r = plist[0] # first cols are all the same
region_p_key = "slk_sidak_p" if "slk_sidak_p" in r else "z_sidak_p" if "z_sidak_p" in r else None
if region_p_key is None:
raise Exception
if float(r[region_p_key]) > region_p:
continue
try:
plist = [
x
for x in plist
if (int(x["start"]) <= int(x["pstart"]) <= int(x["pend"]))
and ((int(x["start"]) <= int(x["pend"]) <= int(x["end"])))
]
except:
print plist
raise
tscores = [float(row["pt"]) for row in plist if "pt" in row]
if max_p:
if any(float(row["p" + p_col_name]) > max_p for row in plist):
continue
ngt05 = sum(1 for row in plist if float(row["p" + p_col_name]) > 0.05)
# logic to try to find t and coef headers and skip if not found
extra_header = []
extra = []
if tscores:
tpos = sum(1 for ts in tscores if ts > 0)
tneg = sum(1 for ts in tscores if ts < 0)
tpn = "%i/%i" % (tpos, tneg)
tsum = str(sum(ts for ts in tscores))
extra_header += ["t.pos/t.neg", "t.sum"]
extra += [tpn, tsum]
else:
tsum = tpn = "NA"
if "p" + coef_col_name not in plist[0] and "pcoefficient" in plist[0]:
coef_col_name = "coefficient"
if "p" + coef_col_name in plist[0]:
coef = sum(float(row["p" + coef_col_name]) for row in plist) / len(plist)
# since we probably had the data logit transformed, here we
# do the inverse and subtract 0.5 since ilogit(0) == 0.5
icoef = (sum(ilogit(float(row["p" + coef_col_name])) for row in plist) / len(plist)) - 0.5
extra_header += ["avg.diff", "ilogit.diff"]
extra += ["%.3f" % coef, "%.3f" % icoef]
else:
coef = icoef = float("nan")
frow = [plist[0][h] for h in rh] + extra
if j == 0:
yield rh + extra_header
j = 1
yield frow
def pipeline(col_num,
step,
dist,
acf_dist,
prefix,
threshold,
seed,
bed_files,
mlog=True,
region_filter_p=1,
region_filter_n=None,
genome_control=False,
db=None,
use_fdr=True):
sys.path.insert(0, op.join(op.dirname(__file__), ".."))
from cpv import acf, slk, fdr, peaks, region_p, stepsize, filter
from cpv._common import genome_control_adjust, genomic_control, bediter
import operator
if step is None:
step = min(acf_dist, stepsize.stepsize(bed_files, col_num))
print >> sys.stderr, "calculated stepsize as: %i" % step
lags = range(1, acf_dist, step)
lags.append(lags[-1] + step)
prefix = prefix.rstrip(".")
putative_acf_vals = acf.acf(bed_files,
lags,
col_num,
simple=False,
mlog=mlog)
acf_vals = []
# go out to max requested distance but stop once an autocorrelation
# < 0.05 is added.
for a in putative_acf_vals:
# a is ((lmin, lmax), (corr, N))
# this heuristic seems to work. stop just above the 0.08 correlation
# lag.
if a[1][0] < 0.04 and len(acf_vals) > 2: break
acf_vals.append(a)
if a[1][0] < 0.04 and len(acf_vals): break
# save the arguments that this was called with.
with open(prefix + ".args.txt", "w") as fh:
print >> fh, " ".join(sys.argv[1:]) + "\n"
import datetime
print >> fh, "date: %s" % datetime.datetime.today()
from .__init__ import __version__
print >> fh, "version:", __version__
with open(prefix + ".acf.txt", "w") as fh:
acf_vals = acf.write_acf(acf_vals, fh)
print >> sys.stderr, "wrote: %s" % fh.name
print >> sys.stderr, "ACF:\n", open(prefix + ".acf.txt").read()
spvals, opvals = [], []
with ts.nopen(prefix + ".slk.bed.gz", "w") as fhslk:
fhslk.write('#chrom\tstart\tend\tp\tregion-p\n')
for row in slk.adjust_pvals(bed_files, col_num, acf_vals):
fhslk.write("%s\t%i\t%i\t%.4g\t%.4g\n" % row)
opvals.append(row[-2])
spvals.append(row[-1])
print >> sys.stderr, "# original lambda: %.2f" % genomic_control(opvals)
del opvals
gc_lambda = genomic_control(spvals)
print >> sys.stderr, "wrote: %s with lambda: %.2f" % (fhslk.name,
gc_lambda)
if genome_control:
fhslk = ts.nopen(prefix + ".slk.gc.bed.gz", "w")
adj = genome_control_adjust(
[d['p'] for d in bediter(prefix + ".slk.bed.gz", -1)])
for i, line in enumerate(ts.nopen(prefix + ".slk.bed.gz")):
print >> fhslk, "%s\t%.5g" % (line.rstrip("\r\n"), adj[i])
fhslk.close()
print >> sys.stderr, "wrote: %s" % fhslk.name
with ts.nopen(prefix + ".fdr.bed.gz", "w") as fh:
fh.write('#chrom\tstart\tend\tp\tregion-p\tregion-q\n')
for bh, l in fdr.fdr(fhslk.name, -1):
fh.write("%s\t%.4g\n" % (l.rstrip("\r\n"), bh))
print >> sys.stderr, "wrote: %s" % fh.name
fregions = prefix + ".regions.bed.gz"
with ts.nopen(fregions, "w") as fh:
list(
peaks.peaks(prefix + ".fdr.bed.gz", -1 if use_fdr else -2,
threshold, seed, dist, fh, operator.le))
n_regions = sum(1 for _ in ts.nopen(fregions))
print >> sys.stderr, "wrote: %s (%i regions)" % (fregions, n_regions)
if n_regions == 0:
sys.exit()
with ts.nopen(prefix + ".regions-p.bed.gz", "w") as fh:
N = 0
fh.write("#chrom\tstart\tend\tmin_p\tn_probes\tz_p\tz_sidak_p\n")
# use -2 for original, uncorrected p-values in slk.bed
for region_line, slk_p, slk_sidak_p, sim_p in region_p.region_p(
prefix + ".slk.bed.gz", prefix + ".regions.bed.gz", -2, step):
fh.write("%s\t%.4g\t%.4g\n" % (region_line, slk_p, slk_sidak_p))
fh.flush()
N += int(slk_sidak_p < 0.05)
print >>sys.stderr, "wrote: %s, (regions with corrected-p < 0.05: %i)" \
% (fh.name, N)
regions_bed = fh.name
header = ts.header(bed_files[0])
#if all(h in header for h in ('t', 'start', 'end')):
if region_filter_n is None: region_filter_n = 0
with ts.nopen(prefix + ".regions-t.bed", "w") as fh:
N = 0
for i, toks in enumerate(
filter.filter(bed_files[0], regions_bed, p_col_name=col_num)):
if i == 0: toks[0] = "#" + toks[0]
else:
if float(toks[6]) > region_filter_p: continue
if int(toks[4]) < region_filter_n: continue
#if region_filter_t and "/" in toks[7]:
# # t-pos/t-neg. if the lower one is > region_filter_t?
# vals = map(int, toks[7].split("/"))
# if min(vals) > region_filter_t: continue
N += 1
print >> fh, "\t".join(toks)
print >>sys.stderr, ("wrote: %s, (regions with region-p "
"< %.3f and n-probes >= %i: %i)") \
% (fh.name, region_filter_p, region_filter_n, N)
try:
from cpv import manhattan
regions = manhattan.read_regions(fh.name)
manhattan.manhattan(prefix + ".slk.bed.gz",
3,
prefix.rstrip(".") + ".manhattan.png",
False, ['#959899', '#484B4C'],
"",
False,
None,
regions=regions,
bonferonni=False)
except ImportError:
pass # they dont have matplotlib
if db is not None:
from cruzdb import Genome
g = Genome(db)
lastf = fh.name
with open(prefix + ".anno.%s.bed" % db, "w") as fh:
fh.write('#')
g.annotate(lastf, ("refGene", "cpgIslandExt"),
out=fh,
feature_strand=True,
parallel=len(spvals) > 500)
print >> sys.stderr, "wrote: %s annotated with %s" % (fh.name, db)
import sys
import toolshed as ts
from sklearn import preprocessing
import numpy as np
regions, weighted = [], []
totlen = 0.0
for d in ts.reader(sys.argv[1]):
totlen += int(d['end']) - int(d['start'])
regions.append(d)
header = ts.header(sys.argv[1])
print "\t".join(header) + "\t" + "weighted_pct"
pct = 100.0
regionlength = 0
opct = regions[0]['resid_pctile']
for d in regions:
regionlength += int(d['end']) - int(d['start'])
if d['resid_pctile'] != opct:
pct -= regionlength / totlen * 100
regionlength = 0
opct = d['resid_pctile']
weighted.append(pct)
X_train = np.array(weighted).reshape(len(weighted), 1)
min_max_scaler = preprocessing.MinMaxScaler(feature_range=(0, 100))
resid_pctile = min_max_scaler.fit_transform(X_train)
for i, d in enumerate(regions):
print "\t".join(d[h] for h in header) + "\t" + "%.9f" % resid_pctile[i]
def filter(p_bed,
region_bed,
max_p=None,
region_p=None,
p_col_name="P.Value",
coef_col_name="logFC"):
ph = ts.header(p_bed)
if (ph[1] + ph[2]).isdigit():
raise Exception('need header in p-value file to run filter')
assert ph[1] == 'start' and ph[2] == 'end' and ph[0] == 'chrom', \
('must have chrom, start, end header for', p_bed)
ph = ['p' + h for h in ph]
rh = ts.header(region_bed)
header = not (rh[1] + rh[2]).isdigit()
if isinstance(p_col_name, str) and p_col_name.isdigit():
p_col_name = int(p_col_name) - 1
if isinstance(p_col_name, (int, long)):
p_col_name = ph[p_col_name][1:]
a = dict(p_bed=p_bed, region_bed=region_bed)
a['p_bed'] = fix_bed(a['p_bed'])
a['header'] = ""
j = 0
for group, plist in groupby(
ts.reader('|bedtools intersect -b %(p_bed)s \
-a %(region_bed)s -wo %(header)s' % a,
header=rh + ph), itemgetter('chrom', 'start', 'end')):
plist = list(plist)
if region_p:
r = plist[0] # first cols are all the same
region_p_key = 'slk_sidak_p' if 'slk_sidak_p' in r \
else 'z_sidak_p' if 'z_sidak_p' in r \
else None
if region_p_key is None: raise Exception
if float(r[region_p_key]) > region_p:
continue
try:
plist = [
x for x in plist
if (int(x['start']) <= int(x['pstart']) <= int(x['pend'])) and
((int(x['start']) <= int(x['pend']) <= int(x['end'])))
]
except:
print(plist)
raise
tscores = [float(row['pt']) for row in plist if 'pt' in row]
if max_p:
if any(float(row['p' + p_col_name]) > max_p for row in plist):
continue
ngt05 = sum(1 for row in plist if float(row['p' + p_col_name]) > 0.05)
# logic to try to find t and coef headers and skip if not found
extra_header = []
extra = []
if tscores:
tpos = sum(1 for ts in tscores if ts > 0)
tneg = sum(1 for ts in tscores if ts < 0)
tpn = "%i/%i" % (tpos, tneg)
tsum = str(sum(ts for ts in tscores))
extra_header += ["t.pos/t.neg", "t.sum"]
extra += [tpn, tsum]
else:
tsum = tpn = "NA"
if 'p' + coef_col_name not in plist[0] and 'pcoefficient' in plist[0]:
coef_col_name = 'coefficient'
if 'p' + coef_col_name in plist[0]:
coef = (sum(float(row['p' + coef_col_name])
for row in plist) / len(plist))
# since we probably had the data logit transformed, here we
# do the inverse and subtract 0.5 since ilogit(0) == 0.5
icoef = (
sum(ilogit(float(row['p' + coef_col_name]))
for row in plist) / len(plist)) - 0.5
extra_header += ["avg.diff", "ilogit.diff"]
extra += ["%.3f" % coef, "%.3f" % icoef]
else:
coef = icoef = float('nan')
frow = [plist[0][h] for h in rh] + extra
if j == 0:
yield rh + extra_header
j = 1
yield frow
def pipeline(col_num, step, dist, acf_dist, prefix, threshold, seed,
bed_files, mlog=True, region_filter_p=1, region_filter_n=None,
genome_control=False, db=None, use_fdr=True):
sys.path.insert(0, op.join(op.dirname(__file__), ".."))
from cpv import acf, slk, fdr, peaks, region_p, stepsize, filter
from cpv._common import genome_control_adjust, genomic_control, bediter
import operator
if step is None:
step = min(acf_dist, stepsize.stepsize(bed_files, col_num))
print >>sys.stderr, "calculated stepsize as: %i" % step
lags = range(1, acf_dist, step)
lags.append(lags[-1] + step)
prefix = prefix.rstrip(".")
putative_acf_vals = acf.acf(bed_files, lags, col_num, simple=False,
mlog=mlog)
acf_vals = []
# go out to max requested distance but stop once an autocorrelation
# < 0.05 is added.
for a in putative_acf_vals:
# a is ((lmin, lmax), (corr, N))
# this heuristic seems to work. stop just above the 0.08 correlation
# lag.
if a[1][0] < 0.04 and len(acf_vals) > 2: break
acf_vals.append(a)
if a[1][0] < 0.04 and len(acf_vals): break
# save the arguments that this was called with.
with open(prefix + ".args.txt", "w") as fh:
print >>fh, " ".join(sys.argv[1:]) + "\n"
import datetime
print >>fh, "date: %s" % datetime.datetime.today()
from .__init__ import __version__
print >>fh, "version:", __version__
with open(prefix + ".acf.txt", "w") as fh:
acf_vals = acf.write_acf(acf_vals, fh)
print >>sys.stderr, "wrote: %s" % fh.name
print >>sys.stderr, "ACF:\n", open(prefix + ".acf.txt").read()
spvals, opvals = [], []
with ts.nopen(prefix + ".slk.bed.gz", "w") as fhslk:
fhslk.write('#chrom\tstart\tend\tp\tregion-p\n')
for row in slk.adjust_pvals(bed_files, col_num, acf_vals):
fhslk.write("%s\t%i\t%i\t%.4g\t%.4g\n" % row)
opvals.append(row[-2])
spvals.append(row[-1])
print >>sys.stderr, "# original lambda: %.2f" % genomic_control(opvals)
del opvals
gc_lambda = genomic_control(spvals)
print >>sys.stderr, "wrote: %s with lambda: %.2f" % (fhslk.name, gc_lambda)
if genome_control:
fhslk = ts.nopen(prefix + ".slk.gc.bed.gz", "w")
adj = genome_control_adjust([d['p'] for d in bediter(prefix + ".slk.bed.gz", -1)])
for i, line in enumerate(ts.nopen(prefix + ".slk.bed.gz")):
print >>fhslk, "%s\t%.5g" % (line.rstrip("\r\n"), adj[i])
fhslk.close()
print >>sys.stderr, "wrote: %s" % fhslk.name
with ts.nopen(prefix + ".fdr.bed.gz", "w") as fh:
fh.write('#chrom\tstart\tend\tp\tregion-p\tregion-q\n')
for bh, l in fdr.fdr(fhslk.name, -1):
fh.write("%s\t%.4g\n" % (l.rstrip("\r\n"), bh))
print >>sys.stderr, "wrote: %s" % fh.name
fregions = prefix + ".regions.bed.gz"
with ts.nopen(fregions, "w") as fh:
list(peaks.peaks(prefix + ".fdr.bed.gz", -1 if use_fdr else -2, threshold, seed,
dist, fh, operator.le))
n_regions = sum(1 for _ in ts.nopen(fregions))
print >>sys.stderr, "wrote: %s (%i regions)" % (fregions, n_regions)
if n_regions == 0:
sys.exit()
with ts.nopen(prefix + ".regions-p.bed.gz", "w") as fh:
N = 0
fh.write("#chrom\tstart\tend\tmin_p\tn_probes\tz_p\tz_sidak_p\n")
# use -2 for original, uncorrected p-values in slk.bed
for region_line, slk_p, slk_sidak_p, sim_p in region_p.region_p(
prefix + ".slk.bed.gz",
prefix + ".regions.bed.gz", -2,
step):
fh.write("%s\t%.4g\t%.4g\n" % (region_line, slk_p, slk_sidak_p))
fh.flush()
N += int(slk_sidak_p < 0.05)
print >>sys.stderr, "wrote: %s, (regions with corrected-p < 0.05: %i)" \
% (fh.name, N)
regions_bed = fh.name
header = ts.header(bed_files[0])
#if all(h in header for h in ('t', 'start', 'end')):
if region_filter_n is None: region_filter_n = 0
with ts.nopen(prefix + ".regions-t.bed", "w") as fh:
N = 0
for i, toks in enumerate(filter.filter(bed_files[0],
regions_bed, p_col_name=col_num)):
if i == 0: toks[0] = "#" + toks[0]
else:
if float(toks[6]) > region_filter_p: continue
if int(toks[4]) < region_filter_n: continue
#if region_filter_t and "/" in toks[7]:
# # t-pos/t-neg. if the lower one is > region_filter_t?
# vals = map(int, toks[7].split("/"))
# if min(vals) > region_filter_t: continue
N += 1
print >>fh, "\t".join(toks)
print >>sys.stderr, ("wrote: %s, (regions with region-p "
"< %.3f and n-probes >= %i: %i)") \
% (fh.name, region_filter_p, region_filter_n, N)
try:
from cpv import manhattan
regions = manhattan.read_regions(fh.name)
manhattan.manhattan(prefix + ".slk.bed.gz", 3, prefix.rstrip(".") + ".manhattan.png",
False, ['#959899', '#484B4C'], "", False, None,
regions=regions, bonferonni=False)
except ImportError:
pass # they dont have matplotlib
if db is not None:
from cruzdb import Genome
g = Genome(db)
lastf = fh.name
with open(prefix + ".anno.%s.bed" % db, "w") as fh:
fh.write('#')
g.annotate(lastf, ("refGene", "cpgIslandExt"), out=fh,
feature_strand=True, parallel=len(spvals) > 500)
print >>sys.stderr, "wrote: %s annotated with %s" % (fh.name, db)
