def test_get_label(self):
''' Test method for getting the index or type label in the settings. '''
self.assertRaises(SettingsError, ElasticSettings.get_label, 'ABC')
self.assertTrue(
isinstance(
ElasticSettings.get_label('MARKER',
idx_type='MARKER',
label='description'), str))
self.assertTrue(isinstance(ElasticSettings.get_label('MARKER'), str))
def filter_queryset(self, request, queryset, view):
''' Override this method to request feature locations. '''
try:
filterable = getattr(view, 'filter_fields', [])
filters = dict([(k, v) for k, v in request.GET.items() if k in filterable])
query_str = filters.get('feature', 'PTPN22')
build = self._get_build(filters.get('build', settings.DEFAULT_BUILD))
if query_str is None or query_str == '':
return [ElasticObject(initial={'error': 'No feature name provided.'})]
search_fields = ['id',
'symbol', 'dbxrefs.ensembl',
'region_name']
sources = ['start', 'stop', 'seqid', 'chromosome',
'disease_loci']
idxs = ElasticSettings.getattr('IDX')
MARKER_IDX = ''
if build == ElasticSettings.get_label('MARKER', label='build'):
MARKER_IDX = 'MARKER'
if MARKER_IDX == '':
for idx in idxs:
if 'MARKER' in idx:
if build == ElasticSettings.get_label(idx, label='build'):
MARKER_IDX = idx
(idx, idx_type) = ElasticSettings.idx_names(MARKER_IDX, 'MARKER')
(idx_r, idx_type_r) = ElasticSettings.idx_names('REGION', 'REGION')
(idx_g, idx_type_g) = ElasticSettings.idx_names('GENE', 'GENE')
idx += ',' + idx_r + ',' + idx_g
idx_type += ',' + idx_type_r + ',' + idx_type_g
equery = BoolQuery(must_arr=Query.query_string(query_str, fields=search_fields))
elastic = Search(search_query=ElasticQuery(equery, sources), size=10, idx=idx, idx_type=idx_type)
docs = elastic.search().docs
locs = []
for doc in docs:
if isinstance(doc, RegionDocument):
doc = Region.pad_region_doc(doc)
loc = doc.get_position(build=build).split(':')
pos = loc[1].replace(',', '').split('-')
locs.append(ElasticObject(
{'feature': query_str,
'chr': loc[0],
'start': int(pos[0]),
'end': int(pos[1]) if len(pos) > 1 else int(pos[0]),
'locusString': query_str+" ("+str(loc[1])+")"}))
return locs
except (TypeError, ValueError, IndexError, ConnectionError):
raise Http404
def _get_marker_build(idx_name):
''' Get the marker build as defined in the settings. '''
try:
idx_key = ElasticSettings.get_idx_key_by_name(idx_name)
return ElasticSettings.get_label(idx_key, label='build')
except (KeyError, SettingsError, TypeError):
logger.error('Marker build not identified from ELASTIC settings.')
return ''
def factory(hit):
''' Factory method for creating specific document object based on
index type of the hit.
@type hit: dict
@param hit: Elasticsearch hit.
'''
(idx, idx_type) = ElasticSettings.get_idx_key_by_name(hit['_index'], idx_type_name=hit['_type'])
if idx is None or idx_type is None:
return PydginDocument(hit)
doc_class = ElasticSettings.get_label(idx, idx_type, label='class')
return doc_class(hit) if doc_class is not None else PydginDocument(hit)
def factory(hit):
''' Factory method for creating types of documents based on
their elasticsearch index type.
@type hit: dict
@param hit: elasticsearch hit.
'''
(idx, idx_type) = ElasticSettings.get_idx_key_by_name(hit['_index'], idx_type_name=hit['_type'])
if idx is None or idx_type is None:
return PydginDocument(hit)
doc_class_str = ElasticSettings.get_label(idx, idx_type, label='class')
doc_class = import_string(doc_class_str) if doc_class_str is not None else None
return doc_class(hit) if doc_class is not None else PydginDocument(hit)
def _build_snp_query(snp_track, chrom, segmin, segmax):
snps = []
snpMeta = {}
maxScore = -1
if snp_track and snp_track != 'None':
# get SNPs based on this segment
mo = re.match(r"(.*)-(.*)", snp_track)
(group, track) = mo.group(1, 2)
try:
snp_track_idx = ElasticSettings.idx('CP_STATS_'+group.upper(), snp_track.upper())
except SettingsError:
snp_track_idx = ElasticSettings.idx('CP_STATS_'+group.upper())+"/"+track
query = ElasticQuery.filtered(Query.terms("seqid", [chrom, str("chr"+chrom)]),
Filter(RangeQuery("end", gte=segmin, lte=segmax)),
utils.snpFields)
snpQuery = Search(search_query=query, search_from=0, size=10000, idx=snp_track_idx)
# snpResult = snpQuery.get_result()
# snps = snpResult['data']
snpResult = snpQuery.get_json_response()
snps = []
for hit in snpResult['hits']['hits']:
snps.append(hit['_source'])
snps = utils.makeRelative(int(segmin), int(segmax), ['start', 'end'], snps)
data_type = ElasticSettings.get_label('CP_STATS_'+group.upper(), None, "data_type")
snpSettings = getattr(chicp_settings, 'STUDY_DEFAULTS').get(data_type)
for s in snps:
if float(s['score']) > maxScore:
maxScore = float(s['score'])
snpSettings['max'] = maxScore
snpMeta = snpSettings
return snps, snpMeta
def _build_snp_query(snp_track, chrom, segmin, segmax):
snps = []
snpMeta = {}
maxScore = -1
if snp_track and snp_track != 'None':
# get SNPs based on this segment
mo = re.match(r"(.*)-(.*)", snp_track)
(group, track) = mo.group(1, 2)
try:
snp_track_idx = ElasticSettings.idx('CP_STATS_'+group.upper(), snp_track.upper())
except SettingsError:
snp_track_idx = ElasticSettings.idx('CP_STATS_'+group.upper())+"/"+track
query = ElasticQuery.filtered(Query.terms("seqid", [chrom, str("chr"+chrom)]),
Filter(RangeQuery("end", gte=segmin, lte=segmax)),
utils.snpFields)
snpQuery = Search(search_query=query, search_from=0, size=10000, idx=snp_track_idx)
# snpResult = snpQuery.get_result()
# snps = snpResult['data']
snpResult = snpQuery.get_json_response()
snps = []
for hit in snpResult['hits']['hits']:
snps.append(hit['_source'])
snps = utils.makeRelative(int(segmin), int(segmax), ['start', 'end'], snps)
data_type = ElasticSettings.get_label('CP_STATS_'+group.upper(), None, "data_type")
snpSettings = getattr(chicp_settings, 'STUDY_DEFAULTS').get(data_type)
for s in snps:
if float(s['score']) > maxScore:
maxScore = float(s['score'])
snpSettings['max'] = maxScore
snpMeta = snpSettings
return snps, snpMeta
def filter_queryset(self, request, queryset, view):
''' Override this method to request feature locations. '''
try:
filterable = getattr(view, 'filter_fields', [])
filters = dict([(k, v) for k, v in request.GET.items() if k in filterable])
ftype = filters.get('ftype').upper()
build = self._get_build(filters.get('build', settings.DEFAULT_BUILD))
if ftype is None or ftype == '':
return [ElasticObject(initial={'error': 'No feature type provided.'})]
idx = 'REGION' if ftype == 'ASSOC_SNP' else ftype
idx_type = 'STUDY_HITS' if idx == 'REGION' else ftype
print(idx_type)
doc_class_str = ElasticSettings.get_label(idx, idx_type, label='class')
doc_class = import_string(doc_class_str) if doc_class_str is not None else None
if ftype == 'REGION':
features = doc_class.get_overlapping_features(build, filters.get('chr').replace('chr', ''),
filters.get('start'), filters.get('end'))
return features
if ftype == 'ASSOC_SNP':
hits = doc_class.get_overlapping_hits(build, filters.get('chr').replace('chr', ''),
filters.get('start'), filters.get('end'))
features = Region.get_immune_snps(hits)
return features
sources = ['start', 'stop', 'seqid', 'chromosome', 'strand',
'biotype', 'giestain', 'name', 'symbol', 'id', 'ref', 'alt']
seqid_param = 'seqid'
end_param = 'stop'
start_param = 'start'
if ftype == 'GENE':
seqid_param = 'chromosome'
elastic = Search.range_overlap_query(filters.get('chr').replace('chr', ''), filters.get('start'),
filters.get('end'), idx=ElasticSettings.idx(idx, idx_type), size=10000,
seqid_param=seqid_param, end_param=end_param, start_param=start_param,
field_list=sources)
docs = elastic.search().docs
features = []
for doc in docs:
if isinstance(doc, RegionDocument):
doc = Region.pad_region_doc(doc)
loc = doc.get_position(build=build).split(':')
pos = loc[1].replace(',', '').split('-')
attributes = {}
feature = {
'name': doc.get_name(),
'id': doc.doc_id(),
'chr': loc[0],
'start': int(pos[0]),
'end': int(pos[1]) if len(pos) > 1 else int(pos[0]),
'strand': doc.get_strand_as_int()
}
if hasattr(doc, "biotype") and getattr(doc, "biotype") is not None:
attributes["biotype"] = getattr(doc, "biotype")
if hasattr(doc, "giestain") and getattr(doc, "giestain") is not None:
attributes["stain"] = getattr(doc, "giestain")
if hasattr(doc, "ref") and hasattr(doc, "alt") and getattr(doc, "ref") is not None and getattr(doc, "alt") is not None:
attributes["alleles"] = getattr(doc, "ref")+"/"+getattr(doc, "alt")
feature['attributes'] = attributes
features.append(ElasticObject(feature))
return features
except (ConnectionError) as err:
logger.error(err)
raise Http404
def chicpea(request):
queryDict = request.GET
user = request.user
context = dict()
context['title'] = 'Capture HiC Plotter'
context['admin_url_path'] = settings.ADMIN_URL_PATH
context['RECAPTCHA_KEY'] = settings.RECAPTCHA_KEY
# context['tissue'] = 'Total_CD4_Activated'
context['searchTerm'] = random.choice(getattr(chicp_settings, 'DEFAULT_GENES'))
if queryDict.get("term"):
context['searchTerm'] = queryDict.get("term")
if queryDict.get("tissue"):
context['tissue'] = queryDict.get("tissue")
else:
context['default_target'] = getattr(chicp_settings, 'DEFAULT_TARGET')
context['default_tissue'] = getattr(chicp_settings, 'DEFAULT_TISSUE')
(idx_keys_auth, idx_type_keys_auth) = get_authenticated_idx_and_idx_types(
user=user, idx_keys=None, idx_type_keys=None)
indexes = list()
tissues = list()
for target in getattr(chicp_settings, 'CP_TARGET'):
if 'CP_TARGET_'+target not in idx_keys_auth:
continue
indexes.append({"value": target, "text": ElasticSettings.get_label('CP_TARGET_'+target)})
elasticJSON = Search(idx=ElasticSettings.idx('CP_TARGET_'+target)).get_mapping(mapping_type="gene_target")
tissueList = list(elasticJSON[ElasticSettings.idx('CP_TARGET_'+target)]
['mappings']['gene_target']['_meta']['tissue_type'].keys())
utils.tissues['CP_TARGET_'+target] = tissueList
tissueList.sort()
for t in tissueList:
tissues.append({"value": t, "text": t.replace("_", " "), "class": target})
context['allIndexes'] = indexes
context['allTissues'] = tissues
snpTracks = OrderedDict()
defaultTrack = getattr(chicp_settings, 'DEFAULT_TRACK')
for group in getattr(chicp_settings, 'CP_STATS'):
if 'CP_STATS_'+group not in idx_keys_auth:
continue
if group == 'UD':
if settings.INCLUDE_USER_UPLOADS:
context['allow_user_uploads'] = True
else:
continue
snp_tracks = list()
for track, details in (ElasticSettings.get_idx_types(idx_name='CP_STATS_'+group, user='******')).items():
if 'CP_STATS_'+group+'.'+track not in idx_type_keys_auth:
continue
snp_tracks.append({"value": track.lower(), "text": details['label']})
if defaultTrack == '':
defaultTrack = track.lower()
snpTracks[ElasticSettings.get_label('CP_STATS_'+group)] = sorted(snp_tracks, key=itemgetter('text'))
context['snpTracks'] = snpTracks
if queryDict.get("snp_track"):
context['snp_track'] = queryDict.get("snp_track")
else:
context['snp_track'] = defaultTrack
return render(request, 'chicp/index.html', context, content_type='text/html')
def chicpea(request):
queryDict = request.GET
user = request.user
context = dict()
context['title'] = 'Capture HiC Plotter'
context['admin_url_path'] = settings.ADMIN_URL_PATH
context['RECAPTCHA_KEY'] = settings.RECAPTCHA_KEY
# context['tissue'] = 'Total_CD4_Activated'
context['searchTerm'] = random.choice(getattr(chicp_settings, 'DEFAULT_GENES'))
if queryDict.get("term"):
context['searchTerm'] = queryDict.get("term")
if queryDict.get("tissue"):
context['tissue'] = queryDict.get("tissue")
else:
context['default_target'] = getattr(chicp_settings, 'DEFAULT_TARGET')
context['default_tissue'] = getattr(chicp_settings, 'DEFAULT_TISSUE')
(idx_keys_auth, idx_type_keys_auth) = get_authenticated_idx_and_idx_types(
user=user, idx_keys=None, idx_type_keys=None)
indexes = list()
tissues = list()
for target in getattr(chicp_settings, 'CP_TARGET'):
if 'CP_TARGET_'+target not in idx_keys_auth:
continue
indexes.append({"value": target, "text": ElasticSettings.get_label('CP_TARGET_'+target)})
elasticJSON = Search(idx=ElasticSettings.idx('CP_TARGET_'+target)).get_mapping(mapping_type="gene_target")
tissueList = list(elasticJSON[ElasticSettings.idx('CP_TARGET_'+target)]
['mappings']['gene_target']['_meta']['tissue_type'].keys())
utils.tissues['CP_TARGET_'+target] = tissueList
tissueList.sort()
for t in tissueList:
tissues.append({"value": t, "text": t.replace("_", " "), "class": target})
context['allIndexes'] = indexes
context['allTissues'] = tissues
snpTracks = OrderedDict()
defaultTrack = getattr(chicp_settings, 'DEFAULT_TRACK')
for group in getattr(chicp_settings, 'CP_STATS'):
if 'CP_STATS_'+group not in idx_keys_auth:
continue
if group == 'UD':
if settings.INCLUDE_USER_UPLOADS:
context['allow_user_uploads'] = True
else:
continue
snp_tracks = list()
for track, details in (ElasticSettings.get_idx_types(idx_name='CP_STATS_'+group, user='******')).items():
if 'CP_STATS_'+group+'.'+track not in idx_type_keys_auth:
continue
snp_tracks.append({"value": track.lower(), "text": details['label']})
if defaultTrack == '':
defaultTrack = track.lower()
snpTracks[ElasticSettings.get_label('CP_STATS_'+group)] = sorted(snp_tracks, key=itemgetter('text'))
context['snpTracks'] = snpTracks
if queryDict.get("snp_track"):
context['snp_track'] = queryDict.get("snp_track")
else:
context['snp_track'] = defaultTrack
return render(request, 'chicp/index.html', context, content_type='text/html')
