def compute_frequency(seq_id, codon_table, showtable):
seq_id = seq_id.rstrip('\n')
if not (seq_id.startswith("ENS") and len(seq_id) == 15):
sys.stderr.write(seq_id + ' is not an Ensembl id' + '\n')
return None
client = ensembl.EnsemblRestClient()
ref = client.get_sequence(seq_id)
sequence = ref['seq']
sequence = sequence.replace('T', 'U')
#for each codon siblings points a list of other codons for the same aa
siblings = dict(
(cod, codgroup) for codgroup, aa in codon_table for cod in codgroup)
cod_count, grp_count, freq = defaultdict(int), defaultdict(int), {}
#counting each codon and each aa
for cod in (sequence[i:i + 3] for i in xrange(0, len(sequence), 3)):
if len(cod) == 3:
cod_count[cod] += 1
grp_count[siblings[cod]] += 1
for cod in siblings.iterkeys(): # the keys of siblings are the 64 codons
if siblings[
cod] in grp_count: #grp_count has value only if aa occurred
freq[cod] = float(cod_count[cod]) / grp_count[siblings[cod]]
freq[siblings[cod]] = float(grp_count[siblings[cod]]) / sum(
grp_count.values())
else:
freq[cod] = '-* Missing *-'
freq[siblings[cod]] = '-* Missing *-'
if showtable == 'true':
print "|-------------|-------------------|-----|------------------|-----------|-----------|---------|\n"+\
"| Amino-acid | AA prob |Codon| Codon prob | All AAs # | This AA # | Codon # |\n"+\
"|-------------|-------------------|-----|------------------|-----------|-----------|---------|\n"
display = '\n'.join('\n'.join(
'%s %-17s %s %-17s %-9s %-9s %s' %
(aa.rjust(13), freq[codgrp], cod.rjust(4), freq[cod],
sum(grp_count.values()), grp_count[siblings[cod]], cod_count[cod])
for i, cod in enumerate(codgrp) if grp_count[siblings[cod]] > 0)
for codgrp, aa in codon_table)
print display
def cod_vr_mapping( pdb_chain_id, vr_filename, transcriptId, alignment ):
# amino acids codes dictionary definition
aaCodes = dict()
aaCodes['G'] = 'Gly' #GLYCINE
aaCodes['P'] = 'Pro' #PROLINE
aaCodes['A'] = 'Ala' #ALANINE
aaCodes['V'] = 'Val' #VALINE
aaCodes['L'] = 'Leu' #LEUCINE
aaCodes['I'] = 'Ile' #ISOLEUCINE
aaCodes['M'] = 'Met' #METHIONINE
aaCodes['C'] = 'Cys' #CYSTEINE
aaCodes['F'] = 'Phe' #PHENYLALANINE
aaCodes['Y'] = 'Tyr' #TYROSINE
aaCodes['W'] = 'Trp' #TRYPTOPHAN
aaCodes['H'] = 'His' #HISTIDINE
aaCodes['K'] = 'Lys' #LYSINE
aaCodes['R'] = 'Arg' #ARGININE
aaCodes['Q'] = 'Gln' #GLUTAMINE
aaCodes['N'] = 'Asn' #ASPARAGINE
aaCodes['E'] = 'Glu' #GLUTAMIC ACID
aaCodes['D'] = 'Asp' #ASPARTIC ACID
aaCodes['S'] = 'Ser' #SERINE
aaCodes['T'] = 'Thr' #THREONINE
client = ensembl.EnsemblRestClient()
ref = client.get_sequence( transcriptId , type="protein")
ensembl_protein_seq = ref['seq']
#print ensembl_protein_seq
#print alignment
ref = client.get_sequence( transcriptId , type="cds")
cod_seq = ref['seq']
cod_seq = cod_seq.replace('T','U')
cod_seq = [cod_seq[i:i+3] for i in range(0, len(cod_seq), 3)]
#print cod_seq
if os.path.exists(vr_filename):
with open(vr_filename) as vr_file:
content = vr_file.readlines()
vr_file.close()
VR = []
for line in content:
VR.append(line.split())
#print VR
i = 0
j = 0
for k in alignment:
#print 'k:' + str(k) +' l:' + ensembl_protein_seq[j]
if j < len(ensembl_protein_seq):
if k == ensembl_protein_seq[j]:
i = i + 1
for row in VR:
if int(row[0]) == i:
#print row
print pdb_chain_id + ' ' + aaCodes[k] + ' ' + cod_seq[j] + ' ' + row[0] + ' ' + row[1] + ' ' + row[2] + ' ' + row[3]
j = j + 1
#pdbId = sys.argv[1]
#parser = PDBParser() #new parser
#pdb = parser.get_structure(pdbId, pdbId + '.pdb')
result = []
return result
import ensembl
import sys
if __name__ == '__main__':
if len(sys.argv) > 3:
species = sys.argv[1]
region = sys.argv[2]
features = sys.argv[3:]
client = ensembl.EnsemblRestClient()
print client.get_region_feature(species, region, features)
else:
print """This is a client for Ensembl REST API resource 'GET overlap/region/:species/:region'
Usage: python get_region_features.py [species] [region] [features]
For example:
python get_region_features.py human 7:140424943-140624564 gene exon
Available features : gene, transcript, cds, exon, repeat, simple, misc,
variation, somatic_variation, structural_variation,
somatic_structural_variation, constrained, regulatory,
segmentation, motif, chipseq, array_probe
"""
def pdb_seq_mapping( pdb_id ):
# amino acids codes dictionary definition
aaCodes = dict()
aaCodes['GLY'] = 'G' #GLYCINE
aaCodes['PRO'] = 'P' #PROLINE
aaCodes['ALA'] = 'A' #ALANINE
aaCodes['VAL'] = 'V' #VALINE
aaCodes['LEU'] = 'L' #LEUCINE
aaCodes['ILE'] = 'I' #ISOLEUCINE
aaCodes['MET'] = 'M' #METHIONINE
aaCodes['CYS'] = 'C' #CYSTEINE
aaCodes['PHE'] = 'F' #PHENYLALANINE
aaCodes['TYR'] = 'Y' #TYROSINE
aaCodes['TRP'] = 'W' #TRYPTOPHAN
aaCodes['HIS'] = 'H' #HISTIDINE
aaCodes['LYS'] = 'K' #LYSINE
aaCodes['ARG'] = 'R' #ARGININE
aaCodes['GLN'] = 'Q' #GLUTAMINE
aaCodes['ASN'] = 'N' #ASPARAGINE
aaCodes['GLU'] = 'E' #GLUTAMIC ACID
aaCodes['ASP'] = 'D' #ASPARTIC ACID
aaCodes['SER'] = 'S' #SERINE
aaCodes['THR'] = 'T' #THREONINE
pdbId = pdb_id
parser = PDBParser() #new parser
pdb = parser.get_structure(pdbId, pdbId + '.pdb')
result = [] #empty list for storing results
processed_chains = []
for chain in pdb.get_chains(): #iterating per each chain
chainId = chain.get_id() #id
if chainId not in processed_chains:
processed_chains.append(chainId)
residuesObj = chain.get_unpacked_list() #residues
chainSeq ='' #getting AA sequence for this chain
#getting AA sequence for this chain
for res in residuesObj:
if res.get_resname() in aaCodes.keys():
chainSeq = chainSeq + aaCodes[res.get_resname()]
#getting uniprotid for this chain
rcsb_url = 'http://www.rcsb.org/pdb/rest/customReport?pdbids=' + pdbId + '.' + chainId + '&customReportColumns=chainId,db_id,db_name&service=wsdisplay&format=text'
headers = { 'User-Agent': 'Mozilla/5.0 (Windows NT 6.0; WOW64; rv:24.0) Gecko/20100101 Firefox/24.0' }
r = requests.get(rcsb_url, headers=headers)
e = ET.ElementTree(ET.fromstring(r.content))
root = e.getroot()
uniprotId = ''
for elem in root.getiterator('dimEntity.db_id'):
uniprotId = elem.text #UniprotId found
if uniprotId == '': #this should never happen
print 'UniprotId not found !'
sys.exit()
#translating uniprotId to Ensembl transcripts
url = 'http://www.uniprot.org/mapping/'
paramsStep2 = {
'from':'ACC',
'to':'ENSEMBL_TRS_ID',
'format':'tab',
'query':uniprotId
}
data = urllib.urlencode(paramsStep2)
request = urllib2.Request(url, data)
user = '******'
request.add_header('User-Agent', user )
try:
response = urllib2.urlopen(request)
except urllib2.HTTPError, err:
if err.code == 503:
response = urllib2.urlopen(request)
page = response.read()
lines = page.split('\n')
transcriptIds = []
for line in lines: #building list of transcripts for uniprotId
if len(line.split('\t')[0]) > 0 and (line.split('\t')[0]) <> 'From':
transcriptIds.append(line.split('\t')[1])
#checking Ensembl reanscriptIds to find one with sequence that is the best match for processed chain
client = ensembl.EnsemblRestClient()
best_score = 0
best_match_tid = ''
best_match_seq = ''
best_alignment = ''
for transcriptId in transcriptIds:
ref = client.get_sequence( transcriptId , type="protein")
sequence = ref['seq']
alns = pairwise2.align.globalxs(chainSeq, sequence, -5, 0)
if alns[0][2] > best_score:
best_score = alns[0][2]
best_match_tid = transcriptId
best_match_seq = sequence
best_alignment = alns[0][0]
result.append([pdbId + '.' + chainId, best_match_tid, best_alignment])