f2 = "/mnt/scratch/endrebak/pyranges_benchmark/data/download/input_15000000.bed.gz"
nrows = int(1e6) # None # int(1e6)
df = pd.read_table(f1, sep="\t", usecols=[0, 1, 2, 5], header=None,
names="Chromosome Start End Strand".split(), nrows=nrows)
df2 = pd.read_table(f2, sep="\t", usecols=[0, 1, 2, 5], header=None,
names="Chromosome Start End Strand".split(), nrows=nrows)
print("Done reading")
start = time()
a = PyRanges(df)
b = PyRanges(df2)
result = a.nearest(b)
end = time()
total = end - start
total_dt = datetime.datetime.fromtimestamp(total)
minutes_seconds = total_dt.strftime('%M\t%S\n')
print(minutes_seconds)
print(result)
# +--------------+----------+----------+----------+
# | Chromosome | Start | End | Strand |
# |--------------+----------+----------+----------|
infile = argv[1]
infile_bg = argv[2]
runs = int(argv[3])
chip_f = infile
background_f = infile_bg
nrows = None
chip = pd.read_table(chip_f,
sep="\t",
nrows=nrows,
usecols=[0, 1, 2, 5],
header=None,
names="Chromosome Start End Strand".split())
background = pd.read_table(background_f,
sep="\t",
nrows=nrows,
usecols=[0, 1, 2, 5],
header=None,
names="Chromosome Start End Strand".split())
cgr = PyRanges(chip)
bgr = PyRanges(background)
for i in range(runs):
print("nearest", i, runs)
result = cgr.nearest(bgr, strandedness="same")