def fidsToSequences(fidlist, config):
''' Given a list of feature IDs, returns a dictionary from FID to its amino acid sequence.
@note Features with no amino acid sequence are discarded.
@param List of feature IDs
@param config Dictionary of configuration variables
@return Dictionary keyed by feature ID of amino acid sequence for feature
'''
cdmi = CDMI_API(config["cdmi_url"])
fidlist = list(set(fidlist))
start = 0
increment = 5000
end = start + increment
counter = len(fidlist)
seqs = {}
while counter > 0:
try:
ps = cdmi.fids_to_protein_sequences(fidlist[start:end])
except HTTPError as e:
if increment > 1:
increment = increment / 2
end = start + increment
sys.stderr.write("caught '%s' error, increment is now %d\n" %(e.reason, increment))
continue
seqs.update(ps)
# Move to next sub-list
start += increment
end += increment
if end >= len(fidlist):
end = len(fidlist)
counter -= increment
return seqs
def genelist2fs(gl):
qid2cds = ids2cds(gl)
fs = {
"description": "Feature set generated by " + ",".join(gl),
"elements": {}
}
cdmie = CDMI_EntityAPI(URLS.cdmi)
cdmic = CDMI_API(URLS.cdmi)
cds_ids = qid2cds.values()
cds2l = cds2locus(cds_ids)
lfunc = cdmic.fids_to_functions(cds2l.values())
fm = cdmie.get_entity_Feature(
cds_ids,
['feature_type', 'source_id', 'sequence_length', 'function', 'alias'])
for i in cds_ids:
if i in fm:
if not fm[i]['function'] and cds2l[i] in lfunc:
fm[i]['function'] = lfunc[cds2l[i]]
fs['elements'][i] = {
"data": {
'type': fm[i]['feature_type'],
'id': i,
'dna_sequence_length': int(fm[i]['sequence_length']),
'function': fm[i]['function'],
'aliases': fm[i]['alias']
}
}
return fs
def subsystemFids(count, config):
''' Query the CDMI for a list of feature IDs in the subsystems.
@param count Number of entities to retrieve in each function call
@param config Dictionary of configuration variables
@return List of subsystem feature IDs
'''
cdmi = CDMI_API(config["cdmi_url"])
cdmi_entity = CDMI_EntityAPI(config["cdmi_url"])
# Get the genes that are in subsystems and in OTUs.
ssdict = dict()
start = 0
done = False
while not done:
subdict = cdmi_entity.all_entities_Subsystem(start, count, ["id"])
ssdict.update(subdict)
start += count
if len(subdict) < count:
done = True
ssids = getFieldFromEntity(ssdict, "id")
sys.stderr.write('Found %d subsystems\n' %(len(ssids)))
# Now lets get a list of FIDs within those subsystems
# Break the complete list into smaller sub-lists to avoid timeouts
start = 0
increment = 10
end = start + increment
counter = len(ssids)
ssfids = []
while counter > 0:
try:
ssfiddict = cdmi.subsystems_to_fids(ssids[start:end], [])
except HTTPError as e:
if increment > 1:
increment = increment / 2
end = start + increment
sys.stderr.write("caught '%s' error, increment is now %d\n" %(e.reason, increment))
continue
for key in ssfiddict:
for ssfid in ssfiddict[key]:
ls = ssfiddict[key][ssfid]
for arr in ls:
if len(arr) > 1:
gl = arr[1]
for l in gl:
ssfids.append(l)
# Move to next sub-list
start += increment
end += increment
if end >= len(ssids):
end = len(ssids)
counter -= increment
# Uniquify!
return list(set(ssfids))
def ids2cds(ql):
cdmic = CDMI_API(URLS.cdmi)
idm = IdMap(URLS.idmap)
gl = set()
rd = {}
eids = []
lids = set()
mids = set()
for gid in ql:
rd[gid] = gid
if 'kb|g.' in gid:
if 'locus' in gid:
lids.add(gid)
elif 'mRNA' in gid:
mids.add(gid)
else:
eids.append(gid)
sid2fids = cdmic.source_ids_to_fids(eids)
for sid in sid2fids:
for fid in sid2fids[sid]:
rd[sid] = fid
if 'locus' in fid:
lids.add(fid)
elif 'mRNA' in fid:
mids.add(fid)
lidmap = ()
if len(lids) > 0: lidmap = idm.longest_cds_from_locus(list(lids))
for lid in lidmap:
for k in lidmap[lid]:
gl.add(k)
midl = list(mids)
midmap = ()
if len(mids) > 0: lidmap = idm.longest_cds_from_mrna(list(mids))
for lid in midmap:
for k in midmap[lid]:
gl.add(k)
for gid in ql:
if 'kb|g.' in gid:
if 'locus' in gid:
for k in lidmap[gid]:
rd[gid] = k
elif 'mRNA' in gid:
for k in midmap[gid]:
rd[gid] = k
else:
if 'locus' in rd[gid]:
for k in lidmap[rd[gid]]:
rd[gid] = k
elif 'mRNA' in rd[gid]:
for k in midmap[rd[gid]]:
rd[gid] = k
return rd
def ids2cds(ql):
cdmic = CDMI_API(URLS.cdmi)
idm = IdMap(URLS.idmap)
gl = set()
rd = {}
eids = []
lids = set()
mids = set()
for gid in ql:
rd[gid] = gid
if 'kb|g.' in gid:
if 'locus' in gid:
lids.add(gid)
elif 'mRNA' in gid:
mids.add(gid)
else:
eids.append(gid)
sid2fids = cdmic.source_ids_to_fids(eids)
for sid in sid2fids:
for fid in sid2fids[sid]:
rd[sid] = fid
if 'locus' in fid:
lids.add(fid)
elif 'mRNA' in fid:
mids.add(fid)
lidmap = ()
if len(lids) > 0: lidmap = idm.longest_cds_from_locus(list(lids))
for lid in lidmap:
for k in lidmap[lid]:
gl.add(k)
midl = list(mids)
midmap = ()
if len(mids) > 0: lidmap = idm.longest_cds_from_mrna(list(mids))
for lid in midmap:
for k in midmap[lid]:
gl.add(k)
for gid in ql:
if 'kb|g.' in gid:
if 'locus' in gid:
for k in lidmap[gid]:
rd[gid] = k
elif 'mRNA' in gid:
for k in midmap[gid]:
rd[gid] = k
else:
if 'locus' in rd[gid]:
for k in lidmap[rd[gid]]:
rd[gid] = k
elif 'mRNA' in rd[gid]:
for k in midmap[rd[gid]]:
rd[gid] = k
return rd
def getGenomefeatures(ref,auth):
gid = [ref]
cdmic = CDMI_API(OTHERURLS.cdmi)
gfids = cdmic.genomes_to_fids(gid,['CDS'])
output = StringIO.StringIO()
for item in gfids.values():
locs = cdmic.fids_to_locations(item)
for key in locs.keys():
# print(locs[key][0][0] + "\t" + str(locs[key][0][1]) + "\t" + str(int(locs[key][0][1]) + int(locs[key][0][3])) + "\t" + locs[key][0][2] + "\t" + key,file = entityfile)
print >>output, locs[key][0][0] + "\t" + str(locs[key][0][1]) + "\t" + str(int(locs[key][0][1]) + int(locs[key][0][3])) + "\t" + locs[key][0][2] + "\t" + key
return output
def getOtuGenomeDictionary(count, config):
''' Obtain a dictionary from OTU representatives to all genomes in the OTU.
@param count Number of entities to retrieve in each function call
@param config Dictionary of configuration variables
@return Dictionary keyed by OTU representative of list of OTU members
'''
cdmi = CDMI_API(config["cdmi_url"])
# Get list of OTUs
otulist = getOtuGenomeIds(count, config)
otudict = cdmi.otu_members(otulist[0])
return otudict
def getDlitFids(count, config):
''' Query the CDMI for a list of feature IDs with direct literature evidence (dlits).
@param count Number of entities to retrieve in each function call
@param config Dictionary of configuration variables
@return List of literature feature IDs
'''
cdmi = CDMI_API(config["cdmi_url"])
cdmi_entity = CDMI_EntityAPI(config["cdmi_url"])
pubdict = dict()
start = 0
done = False
while not done:
subdict = cdmi_entity.all_entities_Publication(start, count, ["id"])
pubdict.update(subdict)
start += count
if len(subdict) < count:
done = True
pubids = getFieldFromEntity(pubdict, "id")
sys.stderr.write("Found %d publication IDs\n" %(len(pubids)))
pub2seq = cdmi_entity.get_relationship_Concerns(pubids, [], [], ["id"])
pubseqs = getFieldFromRelationship(pub2seq, "id", "to")
sys.stderr.write("Found %d protein sequences from publications\n" %(len(pubseqs)))
seq2fids = cdmi_entity.get_relationship_IsProteinFor(pubseqs, [], [], ["id"])
fids = getFieldFromRelationship(seq2fids, "id", "to")
return fids
def fidsToRoles(fidlist, config):
''' Given a list of feature IDs return a dictionary from FID to the list of roles the encoding gene
performs and a dictionary from roles to the FIDs performing them.
@param fidlist List of feature IDs
@param config Dictionary of configuration variables
@return Dictionary keyed by feature ID of list of roles encoding gene performs, dictionary
keyed by role of list of feature IDs performing the role
'''
cdmi = CDMI_API(config["cdmi_url"])
cdmi_entity = CDMI_EntityAPI(config["cdmi_url"])
# Break the complete list into smaller sub-lists to avoid timeouts
start = 0
increment = 1000
end = start + increment
counter = len(fidlist)
fidsToRoles = {}
rolesToFids = {}
while counter > 0:
try:
roledict = cdmi_entity.get_relationship_HasFunctional(fidlist[start:end], [], [], ["id"])
except HTTPError as e:
if increment > 1:
increment = increment / 2
end = start + increment
sys.stderr.write("caught '%s' error, increment is now %d\n" %(e.reason, increment))
continue
flist = getFieldFromRelationship(roledict, "from_link", "rel")
rolelist = getFieldFromRelationship(roledict, "id", "to")
for ii in range(len(flist)):
# We have to use sets here because a bug(?) in get_relationship_HasFunctional allows multiple identical
# links between fids and roles.
# See for example what happens when you call it on g.9647.peg.2332
if flist[ii] in fidsToRoles:
fidsToRoles[flist[ii]].add(rolelist[ii])
else:
fidsToRoles[flist[ii]] = set([rolelist[ii]])
if rolelist[ii] in rolesToFids:
rolesToFids[rolelist[ii]].add(flist[ii])
else:
rolesToFids[rolelist[ii]] = set([flist[ii]])
# Move to next sub-list
start += increment
end += increment
if end >= len(fidlist):
end = len(fidlist)
counter -= increment
# Convert back to lists to not break other functions.
for f in fidsToRoles:
fidsToRoles[f] = list(fidsToRoles[f])
for r in rolesToFids:
rolesToFids[r] = list(rolesToFids[r])
return fidsToRoles, rolesToFids
def genelist2fs(gl):
qid2cds = ids2cds(gl)
fs = {"description" : "Feature set generated by " + ",".join(gl),
"elements" : {}
}
cdmie = CDMI_EntityAPI(URLS.cdmi)
cdmic = CDMI_API(URLS.cdmi)
cds_ids = qid2cds.values()
cds2l = cds2locus(cds_ids);
lfunc = cdmic.fids_to_functions(cds2l.values())
fm = cdmie.get_entity_Feature(cds_ids,['feature_type', 'source_id', 'sequence_length', 'function', 'alias'])
for i in cds_ids:
if i in fm:
if not fm[i]['function'] and cds2l[i] in lfunc:
fm[i]['function'] = lfunc[cds2l[i]]
fs['elements'][i] = {"data" : { 'type' : fm[i]['feature_type'], 'id' : i, 'dna_sequence_length' : int(fm[i]['sequence_length']), 'function' : fm[i]['function'], 'aliases' : fm[i]['alias']}}
return fs
def go_anno_net(meth, net_obj_id=None):
"""Add Gene Ontology annotation to network gene nodes
:param net_obj_id: Network object id
:type net_obj_id: kbtypes.KBaseNetworks.Network
:return: Workspace id
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 5
meth.advance("Prepare annotation service")
#gc = GWAS(URLS.gwas, token=meth.token)
# load from current or other workspace
wsid = meth.workspace_id
# save to current workspace
ws_save_id = meth.workspace_id
meth.advance("Load network object")
wsd = Workspace2(token=meth.token, wsid=wsid)
oc = Ontology(url=URLS.ontology)
net_object = wsd.get(net_obj_id)
nc = Node(net_object['nodes'], net_object['edges'])
idc = IDServerAPI(URLS.ids)
cdmic = CDMI_API(URLS.cdmi)
cdmie = CDMI_EntityAPI(URLS.cdmi)
#idm = IdMap(URLS.idmap)
gids = [
i for i in sorted(nc.ugids.keys()) if 'CDS' in i or 'locus' in i or (
not 'clst' in i and not i.startswith('cluster') and 'ps.' not in i)
]
meth.advance("Get relationships from central data model")
#eids = idc.kbase_ids_to_external_ids(gids)
eids = kb_id2ext_id(idc, gids, 100)
gids2cds = ids2cds(gids)
cgids = gids2cds.values()
cds2l = cds2locus(cgids)
#mrnas_l = cdmie.get_relationship_Encompasses(gids, [], ['to_link'], [])
#mrnas = dict((i[1]['from_link'], i[1]['to_link']) for i in mrnas_l)
#locus_l = cdmie.get_relationship_Encompasses(mrnas.values(), [], ['to_link'], [])
#locus = dict((i[1]['from_link'], i[1]['to_link']) for i in locus_l)
#lgids = [locus[mrnas[i]] for i in gids if i in mrnas.keys()] # ignore original locus ids in gids
lgids = cds2l.values()
meth.advance("Annotate ({:d} nodes, {:d} edges)".format(
len(net_object['nodes']), len(net_object['edges'])))
#ots = oc.get_goidlist(lgids, ['biological_process'], ['IEA'])
ots = oc.get_goidlist(cgids, [], [])
oan = () #oc.get_go_annotation(lgids)
funcs = cdmic.fids_to_functions(lgids)
funcs_org = cdmic.fids_to_functions(cgids)
annotate_nodes(net_object,
ots=ots,
oan=oan,
funcs=funcs,
funcs_org=funcs_org,
eids=eids,
gids2cds=gids2cds,
cds2l=cds2l)
meth.advance("Save annotated object to workspace {}".format(ws_save_id))
obj = {
'type': 'KBaseNetworks.Network',
'data': net_object,
'name': net_obj_id + ".ano",
'meta': {
'original': net_obj_id
}
}
wsd.save_objects({'workspace': ws_save_id, 'objects': [obj]})
return _workspace_output(net_obj_id + ".ano")
def featureset_go_anal(meth,
feature_set_id=None,
p_value=0.05,
ec='IEA',
domain='biological_process',
out_id=None):
"""This method annotate GO terms and execute GO enrichment test
:param feature_set_id: FeatureSet workspace object id
:type feature_set_id: kbtypes.KBaseSearch.FeatureSet
:param p_value: p-value cutoff
:type p_value: kbtypes.Unicode
:param ec: Evidence code list (comma separated, IEA,ISS,IDA,IEP,IPI,RCA ..)
:type ec:kbtypes.Unicode
:param domain: Domain list (comma separated, biological_process,molecular_function,cellular_component)
:type domain: kbtypes.Unicode
:param out_id: Output FeatureSet object identifier
:type out_id: kbtypes.KBaseSearch.FeatureSet
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: GeneTableWidget
"""
meth.stages = 4
meth.advance("Prepare Enrichment Test")
oc = Ontology(url=URLS.ontology)
ws = Workspace2(token=meth.token, wsid=meth.workspace_id)
fs = ws.get(feature_set_id)
qid2cds = ids2cds(fs['elements'].keys())
cds2l = cds2locus(qid2cds.values())
cdmic = CDMI_API(URLS.cdmi)
lfunc = cdmic.fids_to_functions(cds2l.values())
meth.advance("Annotate GO Term")
ec = ec.replace(" ", "")
domain = domain.replace(" ", "")
ec_list = [i for i in ec.split(',')]
domain_list = [i for i in domain.split(',')]
ots = oc.get_goidlist(list(set(qid2cds.values())), domain_list, ec_list)
go_key = lambda go, i, ext: "go.{}.{:d}.{}".format(go, i, ext)
go2cds = {}
for gid in fs['elements']:
lid = qid2cds[gid]
if 'data' in fs['elements'][gid]:
if not fs['elements'][gid]['data']['function']:
fs['elements'][gid]['data']['function'] = lfunc[cds2l[lid]]
if 'metadata' not in fs['elements'][gid]:
fs['elements'][gid]['metadata'] = {}
if lid in ots:
go_enr_list = []
for lcnt, go in enumerate(ots[lid].keys()):
if go not in go2cds: go2cds[go] = set()
go2cds[go].add(lid)
for i, goen in enumerate(ots[lid][go]):
for ext in "domain", "ec", "desc":
fs['elements'][gid]['metadata'][go_key(
go, i, ext)] = goen[ext]
fs['elements'][gid]['metadata'][go_key(
go, i, ext)] = goen[ext]
meth.advance("Execute Enrichment Test")
enr_list = oc.get_go_enrichment(list(set(qid2cds.values())), domain_list,
ec_list, 'hypergeometric', 'GO')
enr_list = sorted(enr_list, key=itemgetter('pvalue'), reverse=False)
header = [
"GO ID", "Description", "Domain", "p-value", "FeatureSet ID (# genes)"
]
fields = []
objects = []
go_enr_smry = ""
for i in range(len(enr_list)):
goen = enr_list[i]
if goen['pvalue'] > float(p_value): continue
cfs = genelist2fs(list(go2cds[goen['goID']]))
goid = goen['goID'].replace(":", "")
fields.append([
goen['goID'], goen['goDesc'][0], goen['goDesc'][1],
"{:12.10f}".format(goen['pvalue']),
"{}_to_{} ({})".format(out_id, goid, len(go2cds[goen['goID']]))
])
objects.append({
'type': 'KBaseSearch.FeatureSet',
'data': cfs,
'name': out_id + "_to_" + goid,
'meta': {
'original': feature_set_id,
'domain': domain,
'ec': ec,
'GO_ID': goen['goID']
}
})
if i < 3:
go_enr_smry += goen['goID'] + "(" + "{:6.4f}".format(
goen['pvalue']) + ")" + goen['goDesc'][0] + "\n"
go_enr_smry
data = {'table': [header] + fields}
meth.advance("Saving output to Workspace")
objects.append({
'type': 'KBaseSearch.FeatureSet',
'data': fs,
'name': out_id,
'meta': {
'original': feature_set_id,
'enr_summary': go_enr_smry
}
})
ws.save_objects({'workspace': meth.workspace_id, 'objects': objects})
return json.dumps(data)
def go_anno_net(meth, net_obj_id=None):
"""Add Gene Ontology annotation to network gene nodes
:param net_obj_id: Network object id
:type net_obj_id: kbtypes.KBaseNetworks.Network
:return: Workspace id
:rtype: kbtypes.Unicode
:output_widget: ValueListWidget
"""
meth.stages = 5
meth.advance("Prepare annotation service")
#gc = GWAS(URLS.gwas, token=meth.token)
# load from current or other workspace
wsid = meth.workspace_id
# save to current workspace
ws_save_id = meth.workspace_id
meth.advance("Load network object")
wsd = Workspace2(token=meth.token, wsid=wsid)
oc = Ontology(url=URLS.ontology)
net_object = wsd.get(net_obj_id)
nc = Node(net_object['nodes'], net_object['edges'])
idc = IDServerAPI(URLS.ids)
cdmic = CDMI_API(URLS.cdmi)
cdmie = CDMI_EntityAPI(URLS.cdmi)
#idm = IdMap(URLS.idmap)
gids = [i for i in sorted(nc.ugids.keys())
if 'CDS' in i or 'locus' in i or (not 'clst' in i and not i.startswith('cluster') and 'ps.' not in i )]
meth.advance("Get relationships from central data model")
#eids = idc.kbase_ids_to_external_ids(gids)
eids = kb_id2ext_id(idc, gids, 100)
gids2cds = ids2cds(gids)
cgids = gids2cds.values()
cds2l = cds2locus(cgids)
#mrnas_l = cdmie.get_relationship_Encompasses(gids, [], ['to_link'], [])
#mrnas = dict((i[1]['from_link'], i[1]['to_link']) for i in mrnas_l)
#locus_l = cdmie.get_relationship_Encompasses(mrnas.values(), [], ['to_link'], [])
#locus = dict((i[1]['from_link'], i[1]['to_link']) for i in locus_l)
#lgids = [locus[mrnas[i]] for i in gids if i in mrnas.keys()] # ignore original locus ids in gids
lgids = cds2l.values()
meth.advance("Annotate ({:d} nodes, {:d} edges)".format(
len(net_object['nodes']), len(net_object['edges'])))
#ots = oc.get_goidlist(lgids, ['biological_process'], ['IEA'])
ots = oc.get_goidlist(cgids, [], [])
oan = () #oc.get_go_annotation(lgids)
funcs = cdmic.fids_to_functions(lgids)
funcs_org = cdmic.fids_to_functions(cgids)
annotate_nodes(net_object, ots=ots, oan=oan, funcs=funcs, funcs_org=funcs_org, eids=eids,
gids2cds=gids2cds, cds2l=cds2l)
meth.advance("Save annotated object to workspace {}".format(ws_save_id))
obj = {
'type': 'KBaseNetworks.Network',
'data': net_object,
'name': net_obj_id + ".ano",
'meta': {
'original': net_obj_id
}
}
wsd.save_objects({'workspace': ws_save_id, 'objects': [obj]})
return _workspace_output(net_obj_id + ".ano")
def featureset_go_anal(meth, feature_set_id=None, p_value=0.05, ec='IEA', domain='biological_process', out_id=None):
"""This method annotate GO terms and execute GO enrichment test
:param feature_set_id: FeatureSet workspace object id
:type feature_set_id: kbtypes.KBaseSearch.FeatureSet
:param p_value: p-value cutoff
:type p_value: kbtypes.Unicode
:param ec: Evidence code list (comma separated, IEA,ISS,IDA,IEP,IPI,RCA ..)
:type ec:kbtypes.Unicode
:param domain: Domain list (comma separated, biological_process,molecular_function,cellular_component)
:type domain: kbtypes.Unicode
:param out_id: Output FeatureSet object identifier
:type out_id: kbtypes.KBaseSearch.FeatureSet
:return: New workspace object
:rtype: kbtypes.Unicode
:output_widget: GeneTableWidget
"""
meth.stages = 4
meth.advance("Prepare Enrichment Test")
oc = Ontology(url=URLS.ontology)
ws = Workspace2(token=meth.token, wsid=meth.workspace_id)
fs = ws.get(feature_set_id)
qid2cds = ids2cds(fs['elements'].keys())
cds2l = cds2locus(qid2cds.values())
cdmic = CDMI_API(URLS.cdmi)
lfunc = cdmic.fids_to_functions(cds2l.values())
meth.advance("Annotate GO Term")
ec = ec.replace(" ","")
domain = domain.replace(" ","")
ec_list = [ i for i in ec.split(',')]
domain_list = [ i for i in domain.split(',')]
ots = oc.get_goidlist(list(set(qid2cds.values())), domain_list, ec_list)
go_key = lambda go, i, ext: "go.{}.{:d}.{}".format(go, i, ext)
go2cds = {}
for gid in fs['elements']:
lid = qid2cds[gid]
if 'data' in fs['elements'][gid]:
if not fs['elements'][gid]['data']['function']: fs['elements'][gid]['data']['function'] = lfunc[cds2l[lid]]
if 'metadata' not in fs['elements'][gid]: fs['elements'][gid]['metadata'] = {}
if lid in ots:
go_enr_list = []
for lcnt, go in enumerate(ots[lid].keys()):
if go not in go2cds: go2cds[go] = set()
go2cds[go].add(lid)
for i, goen in enumerate(ots[lid][go]):
for ext in "domain", "ec", "desc":
fs['elements'][gid]['metadata'][go_key(go, i, ext)] = goen[ext]
fs['elements'][gid]['metadata'][go_key(go, i, ext)] = goen[ext]
meth.advance("Execute Enrichment Test")
enr_list = oc.get_go_enrichment(list(set(qid2cds.values())), domain_list, ec_list, 'hypergeometric', 'GO')
enr_list = sorted(enr_list, key=itemgetter('pvalue'), reverse=False)
header = ["GO ID", "Description", "Domain", "p-value", "FeatureSet ID (# genes)"]
fields = []
objects = []
go_enr_smry = ""
for i in range(len(enr_list)):
goen = enr_list[i]
if goen['pvalue'] > float(p_value) : continue
cfs = genelist2fs(list(go2cds[goen['goID']]))
goid = goen['goID'].replace(":","")
fields.append([goen['goID'], goen['goDesc'][0], goen['goDesc'][1], "{:12.10f}".format(goen['pvalue']), "{}_to_{} ({})".format(out_id, goid,len(go2cds[goen['goID']])) ])
objects.append({'type' : 'KBaseSearch.FeatureSet', 'data' : cfs, 'name' : out_id + "_to_" + goid, 'meta' : {'original' : feature_set_id, 'domain' : domain, 'ec' : ec, 'GO_ID' :goen['goID']}})
if i < 3 :
go_enr_smry += goen['goID']+"(" + "{:6.4f}".format(goen['pvalue']) + ")" + goen['goDesc'][0] + "\n"
go_enr_smry
data = {'table': [header] + fields}
meth.advance("Saving output to Workspace")
objects.append({'type' : 'KBaseSearch.FeatureSet', 'data' : fs, 'name' : out_id, 'meta' : {'original' : feature_set_id, 'enr_summary' : go_enr_smry}})
ws.save_objects({'workspace' : meth.workspace_id, 'objects' :objects})
return json.dumps(data)
