def get_ClaRNA_output_from_dot_bracket(ss, temp=True, verbose=False):
"""
Get dummy ClaRNA output out of dat bracket secondary structure (ss)
Args:
ss (string): secondary structure
Return:
a filename to ClaRNA output"""
from rna_tools.SecondaryStructure import parse_vienna_to_pairs
if ss.find(':') > -1:
chain, ss = ss.split(':')
else:
chain = 'A'
ss = ss
pairs, pairs_pk = parse_vienna_to_pairs(ss, remove_gaps_in_ss=False)
pairs += pairs_pk
txt = 'Classifier: Clarna\n'
txt += 'chains: A 1 ' + str(len(ss)) + '\n'
for bp in pairs:
txt += '%s %i %s %i bp G C WW_cis 1 \n' % (
chain, bp[0], chain, bp[1])
if verbose: print(txt.strip())
if temp:
f = tempfile.NamedTemporaryFile()
name = f.name
else:
name = 'target'
foutCR = name + '.pdb.outCR'
if verbose: print(foutCR)
ft = open(foutCR, 'w')
ft.write(txt)
ft.close()
return foutCR
def get_ClaRNA_output_from_dot_bracket(ss, temp=True, verbose=False):
"""
Get dummy ClaRNA output out of dat bracket secondary structure (ss)
Args:
ss (string): secondary structure
Return:
a filename to ClaRNA output"""
from rna_tools.SecondaryStructure import parse_vienna_to_pairs
if ss.find(':') > -1:
chain,ss = ss.split(':')
else:
chain = 'A'
ss = ss
pairs, pairs_pk = parse_vienna_to_pairs(ss, remove_gaps_in_ss=False)
pairs += pairs_pk
txt = 'Classifier: Clarna\n'
txt += 'chains: A 1 ' + str(len(ss)) + '\n'
for bp in pairs:
txt += '%s %i %s %i bp G C WW_cis 1 \n' % (chain, bp[0], chain, bp[1])
if verbose: print(txt.strip())
if temp:
f = tempfile.NamedTemporaryFile()
name = f.name
else:
name = 'target'
foutCR = name + '.pdb.outCR'
if verbose: print(foutCR)
ft = open(foutCR, 'w')
ft.write(txt)
ft.close()
return foutCR
def get_parser():
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.RawDescriptionHelpFormatter)
parser.add_argument('file', help="file in the Fasta format")
parser.add_argument('--offset', help="offset", type=int)
parser.add_argument("-v", "--verbose", action='count', help="be verbose")
return parser
if __name__ == '__main__':
parser = get_parser()
args = parser.parse_args()
id, seq, ss = open(args.file).read().strip().split('\n')
pairs, pairs_pk = parse_vienna_to_pairs(ss)
if args.offset:
npairs = []
for pair in pairs:
npairs.append([pair[0] + args.offset, pair[1] + args.offset])
for pair in pairs_pk:
npairs.append([pair[0] + args.offset, pair[1] + args.offset])
else:
npairs = pairs + pairs_pk
npairs.sort()
print(npairs)
def rna_dca_mapping(seqfn, gseqfn, file_interactions, noss, noshort, offset, mss, verbose):
"""This function is deviede into
.. warning:: in the line that we load parameters, watch for sep argument that defines seperator of your file (line21)
"""
v = verbose
#
# Load the data, open files and parse information
#
# final ss and seq, ungapped
f = open(seqfn)
header = f.readline().strip() # get rid of header
seq = f.readline().strip()
ss = f.readline().strip()
# gapped
f = open(gseqfn)
header = f.readline().strip() # get rid of header
gseq = f.readline().strip()
gss = f.readline().strip()
# DCA
df = pd.read_csv(file_interactions,sep=" ")
interactions = zip(df['i'].tolist(), df['j'].tolist())
#
# Show input
#
# [(38, 51), (7, 110), (37, 52) from the input
interactions.sort()
print 'interactions:\n', interactions
#
# Process unmapped scores on gaped sequence
# I panel
#
pairs = parse_vienna_to_pairs(gss)[0]
print 'pairs', pairs
print 'UNMAPPED SCORES ' + '/' * len(gseq)
print '123456789112345678921234567893123456789412345678951234567896123456789712345678981234567899123456789'
print gseq
print gss
for i in interactions:
# form 0 or from 1 ?! ## be careful here! scores starts from 0 or 1 !?
ij = [0,0]
ij[0] = i[0] + 1
ij[1] = i[1] + 1
i = ij
if noshort:
delta = i[1] - i[0]
if delta < 6:
continue
# remove interactions from gaps
if gseq[i[0] - 1] == '-' or gseq[i[1] - 1] == '-':
continue
if noss:
if i in pairs:
continue
line_new = 'x'.rjust(i[0]) + 'x'.rjust(i[1] - i[0]) + str(i).rjust(len(gseq) - i[1] + 10)
print line_new
#
# How this mapping works, kurwa?
# II panel
#
print 'MAPPED SCORES //' + '/' * len(gseq)
mapped_interactions = []
for i in interactions:
ij = i
#ij = [0,0]
#ij[0] = i[0] - 1
#ij[1] = i[1] - 1
if v: print 'ij:', ij
if v: print gseq[ij[0]], gseq[ij[1]]
# ok, here I test if this pair comes from gaps, . -> -
if gseq[ij[0]] == '-' or gseq[ij[1]] == '-':
if v: print 'Removed Interaction:', ij
else:
[a,b]=[ij[0] - gseq[:ij[0]].count('-') +1, ij[1] - gseq[:ij[1]].count('-') + 1,]
if v: print i, '->', a, b
mapped_interactions.append([a, b])
print 'Mapped Interactions:\n', mapped_interactions
if v:
for i in mapped_interactions:
print str(i)
pairs = parse_vienna_to_pairs(gss)[0]
print 'pairs', pairs
print '123456789112345678921234567893123456789412345678951234567896123456789712345678981234567899123456789'
print gseq.replace('-','') # what is the gap character, - or . ?
print gss.replace('-', '')
mapped_interactions.sort()
filtered_interactions = []
pairs = parse_vienna_to_pairs(gss.replace('-', ''))[0]
for i in mapped_interactions:
if noshort:
delta = i[1] - i[0]
if delta < 6:
continue
# remove if there is a gap
if seq[i[0] - 1] == '-' or seq[i[1] - 1] == '-':
continue
if noss:
if i in pairs:
continue
line_new = 'x'.rjust(i[0]) + 'x'.rjust(i[1] - i[0]) + str(i).rjust(len(seq) - i[1] + 10)
print line_new
#
# How to include a gap in the mapping?
# III panel, the final
#
print seq
print ss
pairs = parse_vienna_to_pairs(ss.replace('-', ''))[0]
def n_lower_chars(string):
""" https://stackoverflow.com/questions/10953189/count-lower-case-characters-in-a-string """
return sum(1 for c in string if c.islower())
print 'FINAL MAPPING //' + '/' * len(seq)
nmapped_interactions = []
for i in mapped_interactions:
ij = i
if v: print 'ij:', ij
if v: print seq[ij[0] - 1], seq[ij[1] - 1]
# gap mapping
a = ij[0] + n_lower_chars(seq[:ij[0]])
b = ij[1] + n_lower_chars(seq[:ij[1]])
if v: print i, '->', a,b
nmapped_interactions.append([a,b])
mapped_interactions = nmapped_interactions
print(seq)
print(ss)
pairs = parse_vienna_to_pairs(ss)[0]
nmapped_interactions = []
for i in mapped_interactions:
if noshort:
delta = i[1] - i[0]
if delta < 6:
continue
if seq[i[0] - 1] == '-' or seq[i[1] - 1] == '-':
continue
if noss:
if i in pairs:
continue
line_new = 'x'.rjust(i[0]) + 'x'.rjust(i[1] - i[0]) + str(i).rjust(len(seq) - i[1] + 10)
nmapped_interactions.append([i[0], i[1]])
print line_new
if mss:
print(ss)
mapped_interactions = nmapped_interactions
if offset:
# e.g. filter_interaction is [[18, 71]], if offset is 10 then it will give you [[28, 81]]
nmapped_interactions = [[x[0] + offset, x[1] + offset] for x in mapped_interactions]
mapped_interactions = nmapped_interactions
print mapped_interactions
print 'draw_dists(' + str(mapped_interactions) + ')'
print 'output file:', file_interactions+"_mapped.csv"
a = pd.DataFrame(list(mapped_interactions), columns=["i","j"])
a.to_csv(file_interactions+"_mapped.csv",sep=" ")
def rna_dca_mapping(seqfn, gseqfn, file_interactions, noss, noshort, offset,
mss, verbose):
"""This function is deviede into
.. warning:: in the line that we load parameters, watch for sep argument that defines seperator of your file (line21)
"""
v = verbose
#
# Load the data, open files and parse information
#
# final ss and seq, ungapped
f = open(seqfn)
header = f.readline().strip() # get rid of header
seq = f.readline().strip()
ss = f.readline().strip()
# gapped
f = open(gseqfn)
header = f.readline().strip() # get rid of header
gseq = f.readline().strip()
gss = f.readline().strip()
# DCA
df = pd.read_csv(file_interactions, sep=" ")
interactions = list(zip(df['i'].tolist(), df['j'].tolist()))
#
# Show input
#
# [(38, 51), (7, 110), (37, 52) from the input
interactions.sort()
print('interactions:\n', interactions)
#
# Process unmapped scores on gaped sequence
# I panel
#
pairs = parse_vienna_to_pairs(gss)[0]
print('pairs', pairs)
print('UNMAPPED SCORES ' + '/' * len(gseq))
print(
'123456789112345678921234567893123456789412345678951234567896123456789712345678981234567899123456789'
)
print(gseq)
print(gss)
for i in interactions:
# form 0 or from 1 ?! ## be careful here! scores starts from 0 or 1 !?
ij = [0, 0]
ij[0] = i[0] + 1
ij[1] = i[1] + 1
i = ij
if noshort:
delta = i[1] - i[0]
if delta < 6:
continue
# remove interactions from gaps
if gseq[i[0] - 1] == '-' or gseq[i[1] - 1] == '-':
continue
if noss:
if i in pairs:
continue
line_new = 'x'.rjust(
i[0]) + 'x'.rjust(i[1] -
i[0]) + str(i).rjust(len(gseq) - i[1] + 10)
print(line_new)
#
# How this mapping works, kurwa?
# II panel
#
print('MAPPED SCORES //' + '/' * len(gseq))
mapped_interactions = []
for i in interactions:
ij = i
#ij = [0,0]
#ij[0] = i[0] - 1
#ij[1] = i[1] - 1
if v: print('ij:', ij)
if v: print(gseq[ij[0]], gseq[ij[1]])
# ok, here I test if this pair comes from gaps, . -> -
if gseq[ij[0]] == '-' or gseq[ij[1]] == '-':
if v: print('Removed Interaction:', ij)
else:
[a, b] = [
ij[0] - gseq[:ij[0]].count('-') + 1,
ij[1] - gseq[:ij[1]].count('-') + 1,
]
if v: print(i, '->', a, b)
mapped_interactions.append([a, b])
print('Mapped Interactions:\n', mapped_interactions)
if v:
for i in mapped_interactions:
print(str(i))
pairs = parse_vienna_to_pairs(gss)[0]
print('pairs', pairs)
print(
'123456789112345678921234567893123456789412345678951234567896123456789712345678981234567899123456789'
)
print(gseq.replace('-', '')) # what is the gap character, - or . ?
print(gss.replace('-', ''))
mapped_interactions.sort()
filtered_interactions = []
pairs = parse_vienna_to_pairs(gss.replace('-', ''))[0]
for i in mapped_interactions:
if noshort:
delta = i[1] - i[0]
if delta < 6:
continue
# remove if there is a gap
if seq[i[0] - 1] == '-' or seq[i[1] - 1] == '-':
continue
if noss:
if i in pairs:
continue
line_new = 'x'.rjust(
i[0]) + 'x'.rjust(i[1] -
i[0]) + str(i).rjust(len(seq) - i[1] + 10)
print(line_new)
#
# How to include a gap in the mapping?
# III panel, the final
#
print(seq)
print(ss)
pairs = parse_vienna_to_pairs(ss.replace('-', ''))[0]
def n_lower_chars(string):
""" https://stackoverflow.com/questions/10953189/count-lower-case-characters-in-a-string """
return sum(1 for c in string if c.islower())
print('FINAL MAPPING //' + '/' * len(seq))
nmapped_interactions = []
for i in mapped_interactions:
ij = i
if v: print('ij:', ij)
if v: print(seq[ij[0] - 1], seq[ij[1] - 1])
# gap mapping
a = ij[0] + n_lower_chars(seq[:ij[0]])
b = ij[1] + n_lower_chars(seq[:ij[1]])
if v: print(i, '->', a, b)
nmapped_interactions.append([a, b])
mapped_interactions = nmapped_interactions
print(seq)
print(ss)
pairs = parse_vienna_to_pairs(ss)[0]
nmapped_interactions = []
for i in mapped_interactions:
if noshort:
delta = i[1] - i[0]
if delta < 6:
continue
if seq[i[0] - 1] == '-' or seq[i[1] - 1] == '-':
continue
if noss:
if i in pairs:
continue
line_new = 'x'.rjust(
i[0]) + 'x'.rjust(i[1] -
i[0]) + str(i).rjust(len(seq) - i[1] + 10)
nmapped_interactions.append([i[0], i[1]])
print(line_new)
if mss:
print(ss)
mapped_interactions = nmapped_interactions
if offset:
# e.g. filter_interaction is [[18, 71]], if offset is 10 then it will give you [[28, 81]]
nmapped_interactions = [[x[0] + offset, x[1] + offset]
for x in mapped_interactions]
mapped_interactions = nmapped_interactions
print(mapped_interactions)
print('draw_dists(' + str(mapped_interactions) + ')')
print('output file:', file_interactions + "_mapped.csv")
a = pd.DataFrame(list(mapped_interactions), columns=["i", "j"])
a.to_csv(file_interactions + "_mapped.csv", sep=" ")
def get_parser():
parser = argparse.ArgumentParser(
description=__doc__, formatter_class=argparse.RawDescriptionHelpFormatter)
parser.add_argument('file', help="file in the Fasta format")
parser.add_argument('--offset', help="offset", type=int)
parser.add_argument("-v", "--verbose", action='count',
help="be verbose")
return parser
if __name__ == '__main__':
parser = get_parser()
args = parser.parse_args()
id, seq, ss = open(args.file).read().strip().split('\n')
pairs, pairs_pk = parse_vienna_to_pairs(ss)
if args.offset:
npairs = []
for pair in pairs:
npairs.append([pair[0] + args.offset, pair[1] + args.offset])
for pair in pairs_pk:
npairs.append([pair[0] + args.offset, pair[1] + args.offset])
else:
npairs = pairs + pairs_pk
npairs.sort()
print(npairs)