def get_mutation_features(map_file, f_mutations_db_location, output_location):
map_file_obj = open(map_file, 'r')
mutations_db = open(f_mutations_db_location, 'r')
write_to = open(output_location, 'w')
uniprot_to_index_to_disease = build_uniprot_to_index_to_disease(mutations_db)
for line in map_file_obj:
tokens = line.split()
asid = tokens[0].split("_")[0]
prot = tokens[1]
sstart = int(tokens[2])
start = int(tokens[3])
end = int(tokens[4])
eend = int(tokens[5])
if prot not in uniprot_to_index_to_disease:
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
else:
c1_count = score_differences(uniprot_to_index_to_disease, prot, sstart, start)
a_count = score_differences(uniprot_to_index_to_disease, prot, start, end)
c2_count = score_differences(uniprot_to_index_to_disease, prot, end, eend)
prot_len = int(line.split("\t")[7].strip())
canonical_absolute = score_differences(uniprot_to_index_to_disease, prot, 1, prot_len)
print >> write_to, tokens[0] + "\t" + prot + "\t" + repr(c1_count) + "\t" + repr(a_count) + "\t" + repr(
c2_count) + "\t" + repr(canonical_absolute)
write_to.close()
def get_uniprot_elm_features(uniprot_exon_indices_location, uniprot_elm_db_location, output_location):
"""
Reads uniprot ELM file and generates ELM features.
:param uniprot_exon_indices_location:
:param uniprot_elm_db_location:
:param output_location:
:return:
"""
read_from = open(uniprot_elm_db_location, 'r')
uniprot_exon_indices = open(uniprot_exon_indices_location, 'r')
write_to = open(output_location, 'w')
uniprot_to_index_to_elm = {}
for line in read_from:
tokens = line.split("\t")
try:
uniprot = tokens[0].strip()
start = int(tokens[2].strip())
end = int(tokens[3].strip())
for i in range(start, end + 1):
if uniprot in uniprot_to_index_to_elm:
uniprot_to_index_to_elm[uniprot][i] = "*"
else:
uniprot_to_index_to_elm[uniprot] = {i: "*"}
except ValueError:
print "Cannot parse: " + line[0:len(line) - 1]
for line in uniprot_exon_indices:
tokens = line.split()
asid = tokens[0].split("_")[0]
prot = tokens[1]
sstart = int(tokens[2])
start = int(tokens[3])
end = int(tokens[4])
eend = int(tokens[5])
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
if not uniprot_to_index_to_elm.has_key(prot):
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
other_absolute = 0
else:
c1_count = score_differences(uniprot_to_index_to_elm, prot, sstart, start)
a_count = score_differences(uniprot_to_index_to_elm, prot, start, end)
c2_count = score_differences(uniprot_to_index_to_elm, prot, end, eend)
protLen = int(line.split("\t")[7].strip())
canonical_absolute = score_differences(uniprot_to_index_to_elm, prot, 1, protLen)
print >> write_to, tokens[0] + "\t" + prot + "\t" + repr(c1_count) + "\t" + repr(a_count) + "\t" + repr(
c2_count) + "\t" + repr(canonical_absolute)
def get_postranscriptional_modification_features(uniprot_exon_indices_location, uniprot_ptm_db_location,
output_file_location):
"""
Reads uniprot PTM file and generates post translational modification site features.
:param uniprot_exon_indices_location:
:param uniprot_ptm_db_location:
:param output_file_location:
:return:
"""
read_from = open(uniprot_ptm_db_location, 'r')
uniprot_exon_indices = open(uniprot_exon_indices_location, 'r')
write_to = open(output_file_location, 'w')
uniprot_to_index_to_ptm = {}
for line in read_from:
tokens = line.split()
try:
uniprot = tokens[0]
index = int(tokens[1])
ptm = tokens[3]
if uniprot_to_index_to_ptm.has_key(uniprot):
uniprot_to_index_to_ptm[uniprot][index] = "*"
else:
uniprot_to_index_to_ptm[uniprot] = {index: "*"}
except ValueError:
print "Cannot parse: " + line[0:len(line) - 1]
for line in uniprot_exon_indices:
tokens = line.split()
asid = tokens[0].split("_")[0]
prot = tokens[1]
sstart = int(tokens[2])
start = int(tokens[3])
end = int(tokens[4])
eend = int(tokens[5])
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
if not uniprot_to_index_to_ptm.has_key(prot):
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
other_absolute = 0
else:
c1_count = score_differences(uniprot_to_index_to_ptm, prot, sstart, start)
a_count = score_differences(uniprot_to_index_to_ptm, prot, start, end)
c2_count = score_differences(uniprot_to_index_to_ptm, prot, end, eend)
prot_len = int(line.split("\t")[7].strip())
canonical_absolute = score_differences(uniprot_to_index_to_ptm, prot, 1, prot_len)
print >> write_to, tokens[0] + "\t" + prot + "\t" + repr(c1_count) + "\t" + repr(a_count) + "\t" + repr(
c2_count) + "\t" + repr(canonical_absolute)
def get_mutation_features(map_file, f_mutations_db_location, output_location):
map_file_obj = open(map_file, 'r')
mutations_db = open(f_mutations_db_location, 'r')
write_to = open(output_location, 'w')
uniprot_to_index_to_disease = build_uniprot_to_index_to_disease(
mutations_db)
for line in map_file_obj:
tokens = line.split()
asid = tokens[0].split("_")[0]
prot = tokens[1]
sstart = int(tokens[2])
start = int(tokens[3])
end = int(tokens[4])
eend = int(tokens[5])
if prot not in uniprot_to_index_to_disease:
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
else:
c1_count = score_differences(uniprot_to_index_to_disease, prot,
sstart, start)
a_count = score_differences(uniprot_to_index_to_disease, prot,
start, end)
c2_count = score_differences(uniprot_to_index_to_disease, prot,
end, eend)
prot_len = int(line.split("\t")[7].strip())
canonical_absolute = score_differences(uniprot_to_index_to_disease,
prot, 1, prot_len)
print >> write_to, tokens[0] + "\t" + prot + "\t" + repr(
c1_count) + "\t" + repr(a_count) + "\t" + repr(
c2_count) + "\t" + repr(canonical_absolute)
write_to.close()
def get_sable_scores(map_file, f_sable_db_location, uniprot_core_output_location):
map_file_obj = open(map_file, 'r')
sable_db_obj = open(f_sable_db_location, 'r')
write_to = open(uniprot_core_output_location, 'w')
uniprot_to_index_to_core = build_uniprot_to_index_to_core(sable_db_obj)
for line in map_file_obj:
tokens = line.split()
asid = tokens[0].split("_")[0]
prot = tokens[1]
sstart = int(tokens[2])
start = int(tokens[3])
end = int(tokens[4])
eend = int(tokens[5])
rough_a_length = int(int(tokens[0].split("_")[-1].split("=")[1]) / 3)
if asid[0] == "I":
rough_a_length = 0
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
if prot in uniprot_to_index_to_core:
c1_count = score_differences(uniprot_to_index_to_core, prot, sstart, start)
a_count = score_differences(uniprot_to_index_to_core, prot, start, end)
c2_count = score_differences(uniprot_to_index_to_core, prot, end, eend)
prot_len = int(line.split("\t")[7].strip())
canonical_absolute = score_differences(uniprot_to_index_to_core, prot, 1, prot_len)
print >> write_to, tokens[0] + "\t" + prot + "\t" + repr(c1_count) + "\t" + repr(a_count) + "\t" + repr(
c2_count) + "\t" + repr(canonical_absolute)
write_to.close()
def get_uniprot_disorder_features(pulse_path, map_file, iupred_output,
canonical_db_location, disorder_read_out_location):
anchors_map = build_anchors_map(map_file)
uniprot_to_index_to_disorder = fetch_feature('uniprot_iupred', anchors_map, canonical_db_location)
uniprot_to_index_to_disorder = build_uniprot_to_index_to_disorder(iupred_output, uniprot_to_index_to_disorder)
write_to = open(disorder_read_out_location, 'w')
uniprot_exon_indices = open(map_file, 'r')
for line in uniprot_exon_indices:
tokens = line.split('\t')
asid = tokens[0] # .split("_")[0]
prot = tokens[1]
sstart = int(tokens[2])
start = int(tokens[3])
end = int(tokens[4])
eend = int(tokens[5])
rough_a_length = int(int(tokens[0].split("_")[-1].split("=")[1]) / 3)
if asid[0] == "I":
rough_a_length = 0
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
other_absolute = 0
other_a = 0
if prot not in uniprot_to_index_to_disorder:
print "Protein not in uniprot_to_index_to_disorder: ", prot
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
other_absolute = 0
other_a = 0
else:
c1_count = score_differences(uniprot_to_index_to_disorder, prot, sstart, start)
a_count = score_differences(uniprot_to_index_to_disorder, prot, start, end)
c2_count = score_differences(uniprot_to_index_to_disorder, prot, end, eend)
prot_len = int(line.split("\t")[7].strip())
other_prot = asid # or token[0] Customize this line if is need it for test [asid+'//'+prot+'-ST']
other_prot_len = int(line.split("\t")[8].strip())
canonical_absolute = score_differences(uniprot_to_index_to_disorder, prot, 1, prot_len)
other_absolute = score_differences(uniprot_to_index_to_disorder, other_prot, 1, other_prot_len)
other_a_end = start + rough_a_length
if other_a_end > other_prot_len:
other_a_end = other_prot_len
other_a = score_differences(uniprot_to_index_to_disorder, other_prot, start, other_a_end)
print >> write_to, tokens[0] + "\t" + prot + "\t" + repr(c1_count) + "\t" + repr(a_count) + "\t" + repr(
c2_count) + "\t" + repr(canonical_absolute) + "\t" + repr(other_absolute) + "\t" + repr(other_a)
write_to.close()
def get_postranscriptional_modification_features(uniprot_exon_indices_location,
uniprot_ptm_db_location,
output_file_location):
"""
Reads uniprot PTM file and generates post translational modification site features.
:param uniprot_exon_indices_location:
:param uniprot_ptm_db_location:
:param output_file_location:
:return:
"""
read_from = open(uniprot_ptm_db_location, 'r')
uniprot_exon_indices = open(uniprot_exon_indices_location, 'r')
write_to = open(output_file_location, 'w')
uniprot_to_index_to_ptm = {}
for line in read_from:
tokens = line.split()
try:
uniprot = tokens[0]
index = int(tokens[1])
ptm = tokens[3]
if uniprot_to_index_to_ptm.has_key(uniprot):
uniprot_to_index_to_ptm[uniprot][index] = "*"
else:
uniprot_to_index_to_ptm[uniprot] = {index: "*"}
except ValueError:
print "Cannot parse: " + line[0:len(line) - 1]
for line in uniprot_exon_indices:
tokens = line.split()
asid = tokens[0].split("_")[0]
prot = tokens[1]
sstart = int(tokens[2])
start = int(tokens[3])
end = int(tokens[4])
eend = int(tokens[5])
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
if not uniprot_to_index_to_ptm.has_key(prot):
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
other_absolute = 0
else:
c1_count = score_differences(uniprot_to_index_to_ptm, prot, sstart,
start)
a_count = score_differences(uniprot_to_index_to_ptm, prot, start,
end)
c2_count = score_differences(uniprot_to_index_to_ptm, prot, end,
eend)
prot_len = int(line.split("\t")[7].strip())
canonical_absolute = score_differences(uniprot_to_index_to_ptm,
prot, 1, prot_len)
print >> write_to, tokens[0] + "\t" + prot + "\t" + repr(
c1_count) + "\t" + repr(a_count) + "\t" + repr(
c2_count) + "\t" + repr(canonical_absolute)
def get_uniprot_disorder_features(pulse_path, map_file, iupred_output,
canonical_db_location,
disorder_read_out_location):
anchors_map = build_anchors_map(map_file)
uniprot_to_index_to_disorder = fetch_feature('uniprot_iupred', anchors_map,
canonical_db_location)
uniprot_to_index_to_disorder = build_uniprot_to_index_to_disorder(
iupred_output, uniprot_to_index_to_disorder)
write_to = open(disorder_read_out_location, 'w')
uniprot_exon_indices = open(map_file, 'r')
for line in uniprot_exon_indices:
tokens = line.split('\t')
asid = tokens[0] # .split("_")[0]
prot = tokens[1]
sstart = int(tokens[2])
start = int(tokens[3])
end = int(tokens[4])
eend = int(tokens[5])
rough_a_length = int(int(tokens[0].split("_")[-1].split("=")[1]) / 3)
if asid[0] == "I":
rough_a_length = 0
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
other_absolute = 0
other_a = 0
if prot not in uniprot_to_index_to_disorder:
print "Protein not in uniprot_to_index_to_disorder: ", prot
c1_count = 0
a_count = 0
c2_count = 0
canonical_absolute = 0
other_absolute = 0
other_a = 0
else:
c1_count = score_differences(uniprot_to_index_to_disorder, prot,
sstart, start)
a_count = score_differences(uniprot_to_index_to_disorder, prot,
start, end)
c2_count = score_differences(uniprot_to_index_to_disorder, prot,
end, eend)
prot_len = int(line.split("\t")[7].strip())
other_prot = asid # or token[0] Customize this line if is need it for test [asid+'//'+prot+'-ST']
other_prot_len = int(line.split("\t")[8].strip())
canonical_absolute = score_differences(
uniprot_to_index_to_disorder, prot, 1, prot_len)
other_absolute = score_differences(uniprot_to_index_to_disorder,
other_prot, 1, other_prot_len)
other_a_end = start + rough_a_length
if other_a_end > other_prot_len:
other_a_end = other_prot_len
other_a = score_differences(uniprot_to_index_to_disorder,
other_prot, start, other_a_end)
print >> write_to, tokens[0] + "\t" + prot + "\t" + repr(
c1_count) + "\t" + repr(a_count) + "\t" + repr(
c2_count) + "\t" + repr(canonical_absolute) + "\t" + repr(
other_absolute) + "\t" + repr(other_a)
write_to.close()
