def import_3Ds(db_host = 'localhost', db_port = 27017, rna3dhub = False, canonical_only = True, annotate = False, limit = 5000):
client = MongoClient(db_host, db_port)
db_name = ""
if rna3dhub:
db_name = "RNA3DHub"
else:
rna3dHub = None
db_name = "PDB"
db = client[db_name]
rnaview = Rnaview()
if not rna3dhub:
pdb = PDB()
query ="""<orgPdbQuery>
<version>head</version>
<queryType>org.pdb.query.simple.ChainTypeQuery</queryType>
<description>Chain Type: there is a Protein and a RNA chain but not any DNA or Hybrid</description>
<containsProtein>N</containsProtein>
<containsDna>N</containsDna>
<containsRna>Y</containsRna>
<containsHybrid>N</containsHybrid>
</orgPdbQuery>"""
pdb_ids = pdb.query(query)
print "%i 3Ds to process"%len(pdb_ids)
for pdb_id in pdb_ids:
if db['tertiaryStructures'].find_one({'source':"db:pdb:%s"%pdb_id}):
continue
print "Recover %s"%pdb_id
for ts in parsers.parse_pdb(pdb.get_entry(pdb_id)):
try:
ss = None
if annotate:
ss, ts = rnaview.annotate(ts, canonical_only = canonical_only)
save(db, ss, ts, pdb_id, limit)
except Exception, e:
print e
print "No annotation for %s"%pdb_id
save(db, None, ts, pdb_id, limit)
def test():
print "Recovering entry 1EHZ from Protein Databank...\n"
pdb = PDB()
tertiary_structures = parse_pdb(pdb.get_entry('1EHZ'))
print "## 3D annotation ##\n"
print "List of base-pairs computed with RNAVIEW:\n"
for ts in tertiary_structures:
secondary_structure, tertiary_structure = Rnaview().annotate(ts)
print secondary_structure_to_base_pairs(secondary_structure, keep_tertiaries = True)
print "\n## 2D prediction ##\n"
for ts in tertiary_structures:
print "RNA sequence from 1EHZ:\n"
print ts.rna.sequence
print "\nList of base-pairs computed with RNAfold (RNA Vienna Package):\n"
print Rnafold().fold(molecule=ts.rna)
save(db, ss, ts, pdb_id, limit)
except Exception, e:
print e
print "No annotation for %s"%pdb_id
else:
pdb = PDB()
rna3dHub = RNA3DHub()
clusters = rna3dHub.get_clusters()
print "%i 3Ds to process"%len(clusters)
for cluster in clusters['pdb-ids']:
if db['tertiaryStructures'].find_one({'source':"db:pdb:%s"%cluster[0]}):
continue
print "Recover %s"%cluster[0] #we use the first pdb_id in the list of ids making a cluster
for ts in parsers.parse_pdb(pdb.get_entry(cluster[0])):
try:
ss = None
if annotate:
ss, ts = rnaview.annotate(ts, canonical_only = canonical_only)
save(db, ss, ts, cluster[0], limit)
except Exception, e:
print e
print "No annotation for %s"%cluster[0]
def save(db, secondary_structure, tertiary_structure, pdbId, limit):
if db['junctions'].count() >= limit:
print "Limit of %i junctions reached"%limit
sys.exit()
except Exception, e:
print e
print "No annotation for %s"%pdb_id
save(db, None, ts, pdb_id, limit)
else:
pdb = PDB()
rna3dHub = RNA3DHub()
clusters = rna3dHub.get_clusters()
print "%i 3Ds to process"%len(clusters)
for cluster in clusters['pdb-ids']:
pdb_id = cluster[0].split('|')[0]
if db['tertiaryStructures'].find_one({'source':"db:pdb:%s"%pdb_id}):
continue
print "Recover %s"%pdb_id #we use the first pdb_id in the list of ids making a cluster
for ts in parsers.parse_pdb(pdb.get_entry(pdb_id)):
try:
ss = None
if annotate:
ss, ts = rnaview.annotate(ts, canonical_only = canonical_only)
save(db, ss, ts, pdb_id, limit)
except Exception, e:
print e
print "No annotation for %s"%pdb_id
save(db, None, ts, pdb_id, limit)
def save(db, secondary_structure, tertiary_structure, pdbId, limit):
if db['junctions'].count() >= limit:
print "Limit of %i junctions reached"%limit
sys.exit()
def post(self):
data = self.get_argument('data', default=None)
tool = self.get_argument('tool', default=None)
version = self.get_argument('version', default=1)
pdbid = self.get_argument('pdbid', default=None)
output = None
result = None
log = {
'path': self.request.uri,
'tool': tool,
'ip': self.request.remote_ip,
'method': self.request.method,
'date': datetime.datetime.now(),
'status': 'running'
}
logs_db['webservices'].insert(log)
if data and data.startswith('>'): #2D prediction
rnas = parse_fasta(data)
result = []
if len(rnas) == 1: #single molecule prediction (MFE,...)
rna = rnas[0]
secondary_structures = []
if tool == 'rnafold':
secondary_structures.append(
base_pairs_to_secondary_structure(
rna,
Rnafold().fold(rna)))
elif tool == 'contrafold':
secondary_structures.append(
base_pairs_to_secondary_structure(
rna,
Contrafold().fold(rna)))
elif tool == 'rnasubopt':
random_sample = int(
self.get_argument('random_sample', default=20))
for _result in Rnasubopt().fold(
rna, random_sample=random_sample):
secondary_structures.append(
base_pairs_to_secondary_structure(rna, _result))
for ss in secondary_structures:
_result = {
'_id': ss._id,
'name': ss.name,
'source': ss.source,
'rna': {
'name': ss.rna.name,
'sequence': ss.rna.sequence,
'source': ss.rna.source,
'_id': ss.rna._id
}
}
helices_descr = []
for helix in ss.helices:
helix_desc = {
'name': helix['name'],
'location': {
'ends': helix['location']
} if version == 1 else helix['location']
}
if helix.has_key('interactions'):
interactions_descr = []
for interaction in helix['interactions']:
interactions_descr.append({
'orientation':
interaction['orientation'],
'edge1':
interaction['edge1'],
'edge2':
interaction['edge2'],
'location': {
'ends': interaction['location']
} if version == 1 else
interaction['location']
})
helix_desc['interactions'] = interactions_descr
helices_descr.append(helix_desc)
_result['helices'] = helices_descr
single_strands_descr = []
for single_strand in ss.single_strands:
single_strands_descr.append({
'name':
single_strand['name'],
'location': {
'ends': single_strand['location']
} if version == 1 else single_strand['location']
})
_result['singleStrands'] = single_strands_descr
tertiary_interactions_descr = []
for tertiary_interaction in ss.tertiary_interactions:
tertiary_interactions_descr.append({
'orientation':
tertiary_interaction['orientation'],
'edge1':
tertiary_interaction['edge1'],
'edge2':
tertiary_interaction['edge2'],
'location': {
'ends': tertiary_interaction['location']
} if version == 1 else
tertiary_interaction['location']
})
_result[
'tertiaryInteractions'] = tertiary_interactions_descr
result.append(_result)
if tool == 'rnafold' or tool == 'contrafold':
logs_db['webservices'].update({'_id': log['_id']}, {
'$set': {
'status': 'done',
'date': datetime.datetime.now()
}
})
self.write(json_encode(result[0]))
else:
logs_db['webservices'].update({'_id': log['_id']}, {
'$set': {
'status': 'done',
'date': datetime.datetime.now()
}
})
self.write(json_encode(result))
elif len(rnas) >= 2: #structural alignment
if tool == 'mlocarna':
aligned_molecules, consensus2D = Mlocarna().align(rnas)
logs_db['webservices'].update({'_id': log['_id']}, {
'$set': {
'status': 'done',
'date': datetime.datetime.now()
}
})
self.write(to_clustalw(consensus2D, aligned_molecules))
elif tool == 'rnalifold' and data and data.startswith(
'CLUSTAL'
): #computation of consensus structure from sequence alignment
logs_db['webservices'].update(
{'_id': log['_id']},
{'$set': {
'status': 'done',
'date': datetime.datetime.now()
}})
self.write(RnaAlifold().align(data))
elif tool == 'rnaview': #3D annotation
from pyrna.db import PDB
rnaview = Rnaview()
if output == 'rnaml':
if pdbid:
self.write(
rnaview.annotate(pdb_content=PDB().get_entry(pdbid),
raw_output=True))
elif data:
self.write(
rnaview.annotate(pdb_content=data, raw_output=True))
else:
if pdbid:
tertiary_structures = parse_pdb(PDB().get_entry(pdbid))
elif data:
tertiary_structures = parse_pdb(data)
result = []
for ts in tertiary_structures:
(ss, ts) = rnaview.annotate(ts, canonical_only=False)
ss.find_junctions()
_2D_descr = {
'_id': ss._id,
'name': ss.name,
'source': ss.source,
'rna': {
'name': ss.rna.name,
'sequence': ss.rna.sequence,
'source': ss.rna.source,
'_id': ss.rna._id
}
}
helices_descr = []
for helix in ss.helices:
helix_desc = {
'name': helix['name'],
'location': {
'ends': helix['location']
} if version == 1 else helix['location']
}
if helix.has_key('interactions'):
interactions_descr = []
for interaction in helix['interactions']:
interactions_descr.append({
'orientation':
interaction['orientation'],
'edge1':
interaction['edge1'],
'edge2':
interaction['edge2'],
'location': {
'ends': interaction['location']
} if version == 1 else
interaction['location']
})
helix_desc['interactions'] = interactions_descr
helices_descr.append(helix_desc)
_2D_descr['helices'] = helices_descr
single_strands_descr = []
for single_strand in ss.single_strands:
single_strands_descr.append({
'name':
single_strand['name'],
'location': {
'ends': single_strand['location']
} if version == 1 else single_strand['location']
})
_2D_descr['singleStrands'] = single_strands_descr
tertiary_interactions_descr = []
for tertiary_interaction in ss.tertiary_interactions:
tertiary_interactions_descr.append({
'orientation':
tertiary_interaction['orientation'],
'edge1':
tertiary_interaction['edge1'],
'edge2':
tertiary_interaction['edge2'],
'location': {
'ends': tertiary_interaction['location']
} if version == 1 else
tertiary_interaction['location']
})
_2D_descr[
'tertiaryInteractions'] = tertiary_interactions_descr
junctions_descr = []
for junction in ss.junctions:
junctions_descr.append({
'description':
junction['description'],
'location':
junction['location']
})
_2D_descr['junctions'] = junctions_descr
_3D_descr = {
'_id': ts._id,
'name': ts.name,
'source': ts.source,
'rna': {
'name': ts.rna.name,
'sequence': ts.rna.sequence,
'source': ts.rna.source,
'_id': ts.rna._id
}
}
residues_descr = {}
keys = []
for k in ts.residues:
keys.append(k)
keys.sort() #the absolute position are sorted
for key in keys:
atoms = ts.residues[key]['atoms']
atoms_descr = []
for atom in atoms:
atoms_descr.append({
'name': atom['name'],
'coords': atom['coords']
})
residues_descr[str(key)] = {'atoms': atoms_descr}
_3D_descr['residues'] = residues_descr
result.append({"2D": _2D_descr, "3D": _3D_descr})
logs_db['webservices'].update({'_id': log['_id']}, {
'$set': {
'status': 'done',
'date': datetime.datetime.now()
}
})
self.write(json_encode(result))
from pyrna.computations import Cmalign, Rnaview
from bson.objectid import ObjectId
import os
pdb = PDB()
cmalign = Cmalign()
rnaview = Rnaview()
rfam = Rfam(cache_dir = "/home/fjossinet/tmp/Rfam")
families_with_structures = rfam.get_families_with_structures()
for index, row in families_with_structures.iterrows():
rfam_id = row['rfam_id']
tertiary_structures = parse_pdb(pdb.get_entry(row['pdb_id']))
reference_rna = None
ts = None
for tertiary_structure in tertiary_structures:
if tertiary_structure.rna.name == row['chain_name']:
ts = tertiary_structure
reference_rna = ts.rna
break
if ts:
secondary_structure, tertiary_structure = rnaview.annotate(tertiary_structure = ts)
rnas, orgs, consensus_2d = cmalign.align([reference_rna], rfam_id = rfam_id, rfam = rfam)
os.mkdir("/home/fjossinet/tmp/%s"%rfam_id)
os.mkdir("/home/fjossinet/tmp/%s/Molecules"%rfam_id)
from pyrna.computations import Cmalign, Rnaview
from bson.objectid import ObjectId
import os
pdb = PDB()
cmalign = Cmalign()
rnaview = Rnaview()
rfam = Rfam(cache_dir="/home/fjossinet/tmp/Rfam")
families_with_structures = rfam.get_families_with_structures()
for index, row in families_with_structures.iterrows():
rfam_id = row['rfam_id']
tertiary_structures = parse_pdb(pdb.get_entry(row['pdb_id']))
reference_rna = None
ts = None
for tertiary_structure in tertiary_structures:
if tertiary_structure.rna.name == row['chain_name']:
ts = tertiary_structure
reference_rna = ts.rna
break
if ts:
secondary_structure, tertiary_structure = rnaview.annotate(
tertiary_structure=ts)
rnas, orgs, consensus_2d = cmalign.align([reference_rna],
rfam_id=rfam_id,
def post(self):
data = self.get_argument('data', default = None)
tool = self.get_argument('tool', default = None)
version = self.get_argument('version', default = 1)
pdbid = self.get_argument('pdbid', default = None)
output = None
result = None
log = {
'path': self.request.uri,
'tool': tool,
'ip': self.request.remote_ip,
'method': self.request.method,
'date': datetime.datetime.now(),
'status': 'running'
}
logs_db['webservices'].insert(log)
if data and data.startswith('>'): #2D prediction
rnas = parse_fasta(data)
result = []
if len(rnas) == 1: #single molecule prediction (MFE,...)
rna = rnas[0]
secondary_structures = []
if tool == 'rnafold':
secondary_structures.append(base_pairs_to_secondary_structure(rna, Rnafold().fold(rna)))
elif tool == 'contrafold':
secondary_structures.append(base_pairs_to_secondary_structure(rna, Contrafold().fold(rna)))
elif tool == 'rnasubopt':
random_sample = int(self.get_argument('random_sample', default = 20))
for _result in Rnasubopt().fold(rna, random_sample = random_sample):
secondary_structures.append(base_pairs_to_secondary_structure(rna, _result))
for ss in secondary_structures:
_result = {
'_id': ss._id,
'name': ss.name,
'source': ss.source,
'rna': {
'name': ss.rna.name,
'sequence': ss.rna.sequence,
'source': ss.rna.source,
'_id': ss.rna._id
}
}
helices_descr = []
for helix in ss.helices:
helix_desc = {
'name': helix['name'],
'location': {'ends': helix['location']} if version == 1 else helix['location']
}
if helix.has_key('interactions'):
interactions_descr = []
for interaction in helix['interactions']:
interactions_descr.append({
'orientation': interaction['orientation'],
'edge1': interaction['edge1'],
'edge2': interaction['edge2'],
'location': {'ends': interaction['location']} if version == 1 else interaction['location']
})
helix_desc['interactions'] = interactions_descr
helices_descr.append(helix_desc)
_result['helices'] = helices_descr
single_strands_descr = []
for single_strand in ss.single_strands:
single_strands_descr.append({
'name': single_strand['name'],
'location': {'ends': single_strand['location']} if version == 1 else single_strand['location']
})
_result['singleStrands'] = single_strands_descr
tertiary_interactions_descr = []
for tertiary_interaction in ss.tertiary_interactions:
tertiary_interactions_descr.append({
'orientation': tertiary_interaction['orientation'],
'edge1': tertiary_interaction['edge1'],
'edge2': tertiary_interaction['edge2'],
'location': {'ends': tertiary_interaction['location']} if version == 1 else tertiary_interaction['location']
})
_result['tertiaryInteractions'] = tertiary_interactions_descr
result.append(_result)
if tool == 'rnafold' or tool == 'contrafold':
logs_db['webservices'].update({ '_id': log['_id'] }, {'$set': { 'status' : 'done', 'date':datetime.datetime.now()}})
self.write(json_encode(result[0]))
else:
logs_db['webservices'].update({ '_id': log['_id'] }, {'$set': { 'status' : 'done', 'date':datetime.datetime.now()}})
self.write(json_encode(result))
elif len(rnas) >= 2: #structural alignment
if tool == 'mlocarna':
aligned_molecules, consensus2D = Mlocarna().align(rnas)
logs_db['webservices'].update({ '_id': log['_id'] }, {'$set': { 'status' : 'done', 'date':datetime.datetime.now()}})
self.write(to_clustalw(consensus2D, aligned_molecules))
elif tool == 'rnalifold' and data and data.startswith('CLUSTAL'): #computation of consensus structure from sequence alignment
logs_db['webservices'].update({ '_id': log['_id'] }, {'$set': { 'status' : 'done', 'date':datetime.datetime.now()}})
self.write(RnaAlifold().align(data))
elif tool == 'rnaview': #3D annotation
from pyrna.db import PDB
rnaview = Rnaview()
if output == 'rnaml':
if pdbid:
self.write(rnaview.annotate(pdb_content = PDB().get_entry(pdbid), raw_output = True))
elif data:
self.write(rnaview.annotate(pdb_content = data, raw_output = True))
else:
if pdbid:
tertiary_structures = parse_pdb(PDB().get_entry(pdbid))
elif data:
tertiary_structures = parse_pdb(data)
result = []
for ts in tertiary_structures:
(ss, ts) = rnaview.annotate(ts, canonical_only = False)
ss.find_junctions()
_2D_descr = {
'_id': ss._id,
'name': ss.name,
'source': ss.source,
'rna': {
'name': ss.rna.name,
'sequence': ss.rna.sequence,
'source': ss.rna.source,
'_id': ss.rna._id
}
}
helices_descr = []
for helix in ss.helices:
helix_desc = {
'name': helix['name'],
'location': {'ends': helix['location']} if version == 1 else helix['location']
}
if helix.has_key('interactions'):
interactions_descr = []
for interaction in helix['interactions']:
interactions_descr.append({
'orientation': interaction['orientation'],
'edge1': interaction['edge1'],
'edge2': interaction['edge2'],
'location': {'ends': interaction['location']} if version == 1 else interaction['location']
})
helix_desc['interactions'] = interactions_descr
helices_descr.append(helix_desc)
_2D_descr['helices'] = helices_descr
single_strands_descr = []
for single_strand in ss.single_strands:
single_strands_descr.append({
'name': single_strand['name'],
'location': {'ends': single_strand['location']} if version == 1 else single_strand['location']
})
_2D_descr['singleStrands'] = single_strands_descr
tertiary_interactions_descr = []
for tertiary_interaction in ss.tertiary_interactions:
tertiary_interactions_descr.append({
'orientation': tertiary_interaction['orientation'],
'edge1': tertiary_interaction['edge1'],
'edge2': tertiary_interaction['edge2'],
'location': {'ends': tertiary_interaction['location']} if version == 1 else tertiary_interaction['location']
})
_2D_descr['tertiaryInteractions'] = tertiary_interactions_descr
junctions_descr = []
for junction in ss.junctions:
junctions_descr.append({
'description': junction['description'],
'location': junction['location']
})
_2D_descr['junctions'] = junctions_descr
_3D_descr = {
'_id': ts._id,
'name': ts.name,
'source': ts.source,
'rna': {
'name': ts.rna.name,
'sequence': ts.rna.sequence,
'source': ts.rna.source,
'_id': ts.rna._id
}
}
residues_descr = {}
keys=[]
for k in ts.residues:
keys.append(k)
keys.sort() #the absolute position are sorted
for key in keys:
atoms = ts.residues[key]['atoms']
atoms_descr = []
for atom in atoms:
atoms_descr.append({
'name': atom['name'],
'coords': atom['coords']
})
residues_descr[str(key)] = {
'atoms': atoms_descr
}
_3D_descr['residues'] = residues_descr
result.append({"2D": _2D_descr, "3D": _3D_descr})
logs_db['webservices'].update({ '_id': log['_id'] }, {'$set': { 'status' : 'done', 'date':datetime.datetime.now()}})
self.write(json_encode(result))