Harvard:Biophysics 101/2007/Notebook:Katie Fifer/2007-2-8
From OpenWetWare
Script
#!/usr/bin/env python
# Katie Fifer
# asst2.py
# 2/7/07
# Description: A script to generate 10,000 strings of 10 random
# coinflips (H or T) and outputs the tally of contiguous (overlapping
# stretches of 2,3,4,5,6,7,8,9, and 10 H's or T's in that set of
# 10,000 10-mers
import random
# set constants
num_strings = 10000
num_flips = 10
max_repeat = 10
all_strings = [ ]
# random number generation
# generate a list of new strings
for i in range(num_strings):
new_string = .join([random.choice(['H','T']) for n in range (num_flips)])
all_strings.append(new_string)
# figure out how many overlapping stretches of H's there are. will do
# this for each string for each substring. in other words will find
# all instances of 'HH' in each of the strings, and then all instances
# of 'HHH' in each of the strings etc.
def analyze (letter):
for i in range(max_repeat):
# generate the substring to search for. the i + 1 is to account
# for the fact that i starts at 0
substr = .join([letter for n in range (i + 1)])
# for each of the strings in the list, find the number of
# instances of the substring just set (overlapping)
total = 0
for j in range(num_strings):
curr_string = all_strings[j]
count = 0
pos = curr_string.find(substr, 0)
while not pos == -1:
count = count + 1
total = total + 1
pos = curr_string.find(substr, pos + 1)
print substr, total
analyze('H')
analyze('T')
Output
- program run to generate 10,000 strings
H 49831 HH 22372 HHH 9860 HHHH 4232 HHHHH 1813 HHHHHH 754 HHHHHHH 313 HHHHHHHH 127 HHHHHHHHH 44 HHHHHHHHHH 8 T 50169 TT 22622 TTT 10065 TTTT 4401 TTTTT 1937 TTTTTT 824 TTTTTTT 341 TTTTTTTT 122 TTTTTTTTT 37 TTTTTTTTTT 7
Testing
- Output for just 5 strings so you can double check by hand
all strings: ['HHTTHTHTTT', 'THTTTTTTTH', 'TTTHHTHTHT', 'HTHTHTHTTH', 'HTTHHHTTTT'] H 19 HH 4 HHH 1 HHHH 0 HHHHH 0 HHHHHH 0 HHHHHHH 0 HHHHHHHH 0 HHHHHHHHH 0 HHHHHHHHHH 0 T 31 TT 16 TTT 9 TTTT 5 TTTTT 3 TTTTTT 2 TTTTTTT 1 TTTTTTTT 0 TTTTTTTTT 0 TTTTTTTTTT 0
