def draw_all_plots(dataDir, outputDir):
fileNames = core.get_only_files(dataDir)
for fileName in fileNames:
y = []
with open(fileName) as readFile:
for line in readFile:
y.append(float(line))
del y[0]
x = [10 * i for i in range(9)]
plt.bar(x, y, width=10)
# plt.show()
plt.savefig(os.path.join(outputDir,
os.path.basename(fileName)))
return
#!/usr/bin/python
""" Splits replication timings of APOBEC-motif in bins with equal number
of points in each bin"""
import numpy as np
from scipy import stats
import sys
import core
if __name__ == '__main__':
if len(sys.argv) != 3:
sys.exit("Usage: {0} motifRepTimeDir outFile".format(sys.argv[0]))
replication_timings = np.array([])
file_names = core.get_only_files(sys.argv[1])
for name in file_names:
with open(name) as fin:
l = np.array(map(float, fin))
l = l[l != -1]
replication_timings = np.concatenate((replication_timings,
l), axis=1)
print "{0}:\n".format(name)
temp = l[l < 0]
print(len(temp))
sys.exit()
borders = np.linspace(0, 1, num=core.BIN_QUANTITY + 1, endpoint=True)
bin_borders = stats.mstats.mquantiles(replication_timings, borders)
with open(sys.argv[2], 'w') as fout:
fout.write('bin_start\tbin_end\motifs\n')
for i in range(len(bin_borders) - 1):
print "Bin #{0}".format(i)
point_number = sum(map(lambda x: bin_borders[i] < x and
)
frequency = 1.0 * attempts_in_bin["mutations"] / attempts_in_bin["motifs"]
attempts_in_bin["frequency"] = pandas.Series(frequency, index=attempts_in_bin.index)
return attempts_in_bin
def estimate_conditional_probability(bin_borders, event_positions):
""" P{event occurred in bin[i] | event occurred}
dataDir - directory with data, should be splitted and normalized """
events_in_bin = core.split_to_bins(event_positions, bin_borders)
conditional_probability = [0] * len(events_in_bin)
number_of_events = sum(events_in_bin)
for i in range(len(events_in_bin)):
conditional_probability[i] = events_in_bin[i] / number_of_events
return conditional_probability
if __name__ == "__main__":
if len(sys.argv) != 4:
sys.exit("Usage: {0} motifRepTimeBins mutationRepTimeDir " "outDir".format(sys.argv[0]))
OUT_DIR = sys.argv[3]
motifs_in_bin = pandas.read_csv(sys.argv[1], sep="\t")
mutation_rep_time_files = core.get_only_files(sys.argv[2])
for mutation_file in mutation_rep_time_files:
with open(mutation_file) as f:
replication_timings = map(float, f)
frequency = calculate_frequency(motifs_in_bin, replication_timings)
outFile = os.path.join(OUT_DIR, os.path.basename(mutation_file))
frequency.to_csv(outFile, sep="\t")
