def Cij(pos1,pos2):
"""
Calculate the Hofacker covariation between two RNA positions
in an alignment. Each position must be converted into a string
of RNA.
Modules required:
- sys
- hammdist
- equLen
Usage: <sequence1> <sequence2>
References:
- Hofacker et al. 2002. Secondary Structure Prediction for
Aligned RNA Sequences. J. Mol. Biol. 319: 1059-1066.
"""
equLen(pos1,pos2)
allow, Cov = ['CG','GC','AU','UA','GU','UG'], 0
for a in range(6):
for b in range(6):
if b>a:
fxy, fyz = 0, 0
for c in range(len(pos1)):
if pos1[c]+pos2[c] == allow[a]: fxy += 1
if pos1[c]+pos2[c] == allow[b]: fyz += 1
hamm = hammDist(allow[a],allow[b])
Cov += (fxy/float(len(pos1)))*hamm*(fyz/float(len(pos2)))
return Cov
def basepairs(seq1, seq2, acid='RNA'):
"""
Return two sequences that represents all the basepairs presents in the sequences inputted
Usage: <sequence 1> <sequence 2>
"""
if acid == 'RNA':
allow = ['AA', 'AC', 'AG', 'AU',
'CA', 'CC', 'CG', 'CU',
'GA', 'GC', 'GG', 'GU',
'UA', 'UC', 'UG', 'UU']
if acid == 'DNA':
allow = ['AA', 'AC', 'AG', 'AT',
'CA', 'CC', 'CG', 'CT',
'GA', 'GC', 'GG', 'GT',
'TA', 'TC', 'TG', 'TT']
total = []
equLen(seq1, seq2)
for index in range(len(seq1)):
if (seq1[index]+seq2[index] in allow) and (seq1[index]+seq2[index] not in total):
total.append(seq1[index]+seq2[index])
pos1, pos2 = '',''
for elem in total:
pos1 += elem[0]
pos2 += elem[1]
return pos1, pos2
def similarity(str1, str2):
"""
Returns the similarity index between two string. Both string must have equal lenght.
Module required:
- equLen (from rnatk.stats)
Usage: <string 1> <string 2>
"""
equLen(str1, str2)
matches = 0.0
for index in range(len(str1)):
if str1[index] == str2[index]:
matches += 1
similarity = matches/len(str1)
return similarity
def hammDist(str1,str2):
"""
Find the Hamming distance between the two strings.
Modules required:
- sys
- equLen
Usage: <string 1> <string 2>
"""
equLen(str1,str2)
diff = 0
for a, b in zip(str1,str2):
if a != b:
diff += 1
return diff
def replaceGapPos_by_consensus(sequence, consensus):
"""
Give a sequence, this function replace the gaps by the correspondent
nucleotide from the consensus sequence.
Module required:
- equLen (rnatk.stats)
Usage: <sequence> <consensus>
"""
equLen(sequence, consensus)
sequence = list(sequence)
for index in range(len(sequence)):
if sequence[index] == '-':
sequence[index] = consensus[index]
return ''.join(sequence)
def fit_to_secondStruct(secStr, sequence):
"""
Given a RNA sequence and a RNA secondary structure in dot-bracket notation, this function try to fit the
structure to the sequence and return such scondary structure in dot-bracket notation.
Module required:
- equLen (from rnatk.stats)
Usage: <secondary structure> <RNA sequence>
"""
equLen(secStr, sequence) # chech if the length of the structure is equal to the sequence
charLst = list(secStr) # build a list where each item is one character of the secondary structure inputted
if charLst.count('(') != charLst.count(')'): # this check if the amount of '(' characters is equal to the amount of
# ')' characters (which it should be!)
print 'A bracket is missing. Quitting'
sys.exit()
index, search = 0, 1
allow = ['CG','GC','AU','UA','GU','UG']
while ('(' in charLst) and (')' in charLst):
repeat = False
if charLst[index] == '(': # this is the reference point (index) for search the next ')' character
if charLst[index+search] == '.' or charLst[index+search] == 'r' or charLst[index+search] == 'l':
search += 1
if charLst[index+search] == '(': # if the next character is this (open-bracket), this is the new reference
# point and the process restart with it.
index += search # New reference point
search = 1 # the search restart to 1
repeat = True # this means that this new reference point must be ignored by the next line
if (charLst[index+search] == ')') and repeat==False:
if (sequence[index]+sequence[index+search]) in allow:
charLst[index] = 'l'
charLst[index+search] = 'r'
else:
charLst[index] = '.'
charLst[index+search] = '.'
index = 0
search = 1
else:
index += 1
result = ''.join(charLst) # here the resulted is compiled to a string variable
return result.replace('l', '(').replace('r', ')')
def entropy(logbase, *seqs):
"""
This function find the entropy for one or several sequences (joint entropy)
Equation used here:
H(X_1, ..., X_n) = -sum{x_1}...sum{x_n} P(x_1, ..., x_n)log[P(x_1, ..., x_n)]
Module required:
- math
Usage: <logbase> <seq1> ... <seq2>
"""
equLen(*seqs)
joint_list = []
for index_A in range(len(seqs[0])):
paired = ''
for index_B in range(len(seqs)):
paired += seqs[index_B][index_A]
joint_list.append(paired)
set_pair = set(joint_list)
Hjoint = 0
for elem in set_pair:
prob = joint_list.count(elem)/float(len(joint_list))
Hjoint -= prob*math.log(prob,logbase)
return Hjoint