def pfam_annotate(data):
data["PFAM_domains"] = "-"
data["Link_to_pfam"] = "-"
for index, row in enumerate(data.iterrows()):
pfam_domains = []
uniprot_id = row[1][18].split(",")
aa_pos = int(row[1][12])
for seq in uniprot_id:
if seq == "-":
continue
else:
pfam_url = "https://pfam.xfam.org/protein/"
full_url = os.path.join(pfam_url, seq)
try:
f = prody.searchPfam(seq)
for i in f.items():
start_pos = int(i[1]["locations"][0]["start"])
end_pos = int(i[1]["locations"][0]["end"])
if aa_pos >= start_pos and aa_pos <= end_pos:
pfam_domains.append(str(i[1]["id"])+":"+\
str(start_pos)+"-"+str(end_pos))
except Exception:
continue
if len(pfam_domains) == 0:
continue
else:
data.loc[index, "PFAM_domains"] = ",".join(pfam_domains)
data.loc[index, "Link_to_pfam"] = full_url
return data
def _searchPfam(self, refresh=False, **kwargs):
assert type(refresh) is bool
if refresh is True or self.Pfam is None:
try:
self.Pfam = searchPfam(self.uniq_acc, **kwargs)
except:
self.Pfam = {}
raise
return self.Pfam
def evol_search(query, **kwargs):
import prody
from os.path import join, split
pfam_results = prody.searchPfam(query, **kwargs)
if pfam_results is None:
return
outname = kwargs.get('outname', None)
delimiter = kwargs.get('delimiter', '\t')
if outname:
folder, outname = split(outname)
filepath = join(prody.utilities.makePath(folder), outname)
out = open(filepath, 'wb')
else:
from sys import stdout as out
title = delimiter.join(['acc', 'id', 'type', 'e-value']) + '\n'
out.write(title)
for key in pfam_results:
val = pfam_results[key]
evalue = ''
for i, location in enumerate(val.get('locations', [])):
temp = location.get('evalue', None)
if temp:
if i == 0:
evalue = float(temp)
else:
if float(temp) < evalue:
evalue = float(temp)
output = delimiter.join([
val.get('accession', ' '),
val.get('id', ' '),
val.get('type', ' '),
str(evalue)
]) + '\n'
out.write(output)
if outname:
prody.LOGGER.info('Search results written in {0}.'.format(filepath))
out.close()
def evol_search(query, **kwargs):
import prody
from os.path import join, split
pfam_results = prody.searchPfam(query, **kwargs)
if pfam_results is None:
return
outname = kwargs.get('outname', None)
delimiter = kwargs.get('delimiter', '\t')
if outname:
folder, outname = split(outname)
filepath = join(prody.utilities.makePath(folder), outname)
out = open(filepath, 'wb')
else:
from sys import stdout as out
title = delimiter.join(['acc', 'id', 'type', 'e-value']) + '\n'
out.write(title)
for key in pfam_results:
val = pfam_results[key]
evalue = ''
for i, location in enumerate(val.get('locations', [])):
temp = location.get('evalue', None)
if temp:
if i==0:
evalue = float(temp)
else:
if float(temp) < evalue:
evalue = float(temp)
output = delimiter.join([val.get('accession', ' '),
val.get('id', ' '),
val.get('type', ' '),
str(evalue)]) + '\n'
out.write(output)
if outname:
prody.LOGGER.info('Search results written in {0}.'.format(filepath))
out.close()
def evol_search(query, **kwargs):
import prody
from os.path import join, split
pfam_results = prody.searchPfam(query, **kwargs)
if pfam_results is None:
return
outname = kwargs.get("outname", None)
delimiter = kwargs.get("delimiter", "\t")
if outname:
folder, outname = split(outname)
filepath = join(prody.utilities.makePath(folder), outname)
out = open(filepath, "wb")
else:
from sys import stdout as out
title = delimiter.join(["acc", "id", "type", "e-value"]) + "\n"
out.write(title)
for key in pfam_results:
val = pfam_results[key]
evalue = ""
for i, location in enumerate(val.get("locations", [])):
temp = location.get("evalue", None)
if temp:
if i == 0:
evalue = float(temp)
else:
if float(temp) < evalue:
evalue = float(temp)
output = (
delimiter.join([val.get("accession", " "), val.get("id", " "), val.get("type", " "), str(evalue)])
+ "\n"
)
out.write(output)
if outname:
prody.LOGGER.info("Search results written in {0}.".format(filepath))
out.close()
if (len(uniprot_code)):
### get the uniprot canonical sequence from the database (full kinase sequence)
raw_sequence = o3i.getCanonicalSequenceFromUniprotCode(uniprot_code)
uniprot_sequence = Seq(raw_sequence, generic_protein)
seq_id = o3i.getCanonicalSeqIdFromUniprotCode(uniprot_code)
# get a list of start and end positions for a given annotation on the sequence
# if none is returned, then the annotation is not found for this particular sequence
cur_domain_idx = 1
#first let's get all domain annotations from PFAM, if available
#
#print(uniprot_code)
try:
pfam_annotations = prody.searchPfam(uniprot_code)
except Exception:
continue
pfam_families = pfam_annotations.keys()
#for pfam_family in pfam_families:
# if pfam_family in expected_pfam_families:
# print(pfam_annotations[pfam_family])
for family in expected_pfam_families:
if (family in pfam_annotations):
subSequence = ""
if (len(pfam_annotations[family]["locations"]) > 1):
# print(uniprot_code,pfam_annotations[family]["locations"])
for loc_idx, location in enumerate(
pfam_annotations[family]["locations"]):