def create_facets(self, queryset):
statuses = OrderedDict([('name', 'status'), ('label', 'Status'),
('items', [])])
organism = OrderedDict([('name', 'organism'), ('label', 'Organism'),
('items', [])])
sequence = OrderedDict([('name', 'divergence'),
('label', 'Divergence'), ('items', [])])
species_set = queryset.species()
for species in species_set:
organism["items"].append({"name": species[0], "label": species[1]})
for query_status in queryset.statuses():
statuses["items"].append({
"name":
Mapping.status_type(query_status),
"label":
Mapping.status_type(query_status).replace("_",
" ").capitalize()
})
differences = queryset.divergences()
if differences[0]:
sequence["items"].append({
"label": "identical",
"name": "identical",
"count": differences[0]
})
if differences[1]:
sequence["items"].append({
"label": "small",
"name": "small",
"count": differences[1]
})
if differences[2]:
sequence["items"].append({
"label": "large",
"name": "large",
"count": differences[2]
})
if len(species_set) == 1:
chromosomes = OrderedDict([('name', 'chromosomes'),
('label', 'Chromosomes'),
('items', [])])
for chromosome in queryset.chromosomes():
chromosomes["items"].append({
'name': chromosome.lower(),
'label': chromosome.upper()
})
return [statuses, organism, sequence, chromosomes]
return [statuses, organism, sequence]
def get_status(mapping):
try:
mapping_status = Mapping.status_type(mapping.status)
except (IndexError, CvUeStatus.DoesNotExist):
# TODO: should log this anomaly or do something else
mapping_status = None
return mapping_status
def create_facets(self, queryset):
statuses = OrderedDict([('name', 'status'), ('label', 'Status'),
('items', [])])
organism = OrderedDict([('name', 'organism'), ('label', 'Organism'),
('items', [])])
sequence = OrderedDict([('name', 'sequence'), ('label', 'Sequence'),
('items', [])])
#######
# TODO
mappings = OrderedDict([('name', 'mappings'), ('label', 'Type'),
('items', [])])
#######
for species in queryset.species():
organism["items"].append({"name": species[0], "label": species[1]})
for status in queryset.statuses():
statuses["items"].append({
"name":
Mapping.status_type(status),
"label":
Mapping.status_type(status).replace("_", " ").capitalize()
})
differences = queryset.divergences()
if differences[0]:
sequence["items"].append({
"label": "identical",
"name": "identical",
"count": differences[0]
})
if differences[1]:
sequence["items"].append({
"label": "small",
"name": "small",
"count": differences[1]
})
if differences[2]:
sequence["items"].append({
"label": "large",
"name": "large",
"count": differences[2]
})
# return [ statuses, organism, sequence, mappings ]
return [statuses, organism, sequence]
def get(self, request):
#
# Mapping stats: general and Uniprot/Ensembl specific
#
mappings_count = Mapping.objects.count() # tot mappings
uniprot_mapped_count = Mapping.objects.values(
'uniprot').distinct().count() # tot mapped uniprot entries
uniprot_not_mapped_sp_count = None # tot non mapped Swiss-Prot entries, NOTE: NO WAY TO GET IT AT THE MOMENT
all_entry_types = dict((entry.id, entry.description)
for entry in CvEntryType.objects.all())
sp_entry_type_ids = [
k for (k, v) in all_entry_types.items()
if v.lower().startswith('swiss')
]
nonsp_entry_type_ids = list(
set(all_entry_types.keys()).difference(sp_entry_type_ids))
#
# NOTE:
# This is not requested but it's computation is reported (commented) here for completeness
uniprot_mapped_sp_count = 0 # tot mapped Swiss-Prot entries, NOTE: NOT REQUESTED BUT HERE FOR COMPLETENESS
# if sp_entry_type_ids:
# query_filter = Q(mapping_history__entry_type=sp_entry_type_ids[0])
#
# for i in range(1, len(sp_entry_type_ids)):
# query_filter = query_filter | Q(mapping_history__entry_type=sp_entry_type_ids[i])
#
# uniprot_mapped_sp_count = Mapping.objects.filter(query_filter).values('uniprot').distinct().count()
#
# tot non mapped genes which none of its transcripts match to any SwissProt entry
gene_not_mapped_sp_count = 0
#
# NOTE:
# Here we're counting genes not mapped to Swiss-Prot entries among the mapped genes.
# As discussed with UniProt, they'd prefer counting among the non mapped genes, which
# in this case coincide with counting the non-mapped genes in general, as done below.
#
# if nonsp_entry_type_ids:
# query_filter = Q(mapping_history__entry_type=nonsp_entry_type_ids[0])
#
# for i in range(1, len(sp_entry_type_ids)):
# query_filter = query_filter | Q(mapping_history__entry_type=nonsp_entry_type_ids[i])
#
# gene_not_mapped_sp_count = Mapping.objects.filter(query_filter).values('transcript__gene').distinct().count()
gene_ids = set(gene.gene_id for gene in EnsemblGene.objects.all())
gene_mapped_ids = set(
item['transcript__gene']
for item in Mapping.objects.values('transcript__gene').distinct())
gene_mapped_count = len(gene_mapped_ids) # tot mapped Ensembl genes
gene_not_mapped_sp_count = len(gene_ids.difference(gene_mapped_ids))
transcript_mapped_count = Mapping.objects.values(
'transcript').distinct().count() # tot mapped Ensembl transcripts
#
# Stats relative to mapping labels
#
all_labels = CvUeLabel.objects.all()
label_counts = []
for label in all_labels:
count = UeMappingLabel.objects.filter(label=label).count()
label_counts.append({'label': label.description, 'count': count})
#
# Stats for mapping status
#
status_counts = []
status_totals = Mapping.objects.values('status').annotate(
total=Count('status'))
for status_count in status_totals:
status_counts.append({
'status':
Mapping.status_type(status_count['status']),
'count':
status_count['total']
})
serializer = MappingStatsSerializer({
'mapping': {
'total': mappings_count,
'uniprot': {
'mapped': uniprot_mapped_count,
'not_mapped_sp': uniprot_not_mapped_sp_count
},
'ensembl': {
'gene_mapped': gene_mapped_count,
'gene_not_mapped_sp': gene_not_mapped_sp_count,
'transcript_mapped': transcript_mapped_count
}
},
'status': status_counts,
'label': label_counts
})
return Response(serializer.data)
def build_mapping(cls, mapping, fetch_sequence=False, authenticated=False):
mapping_history = mapping.mapping_history.select_related(
'release_mapping_history').select_related(
'release_mapping_history__ensembl_species_history').latest(
'mapping_history_id')
release_mapping_history = mapping_history.release_mapping_history
ensembl_history = mapping_history.release_mapping_history.ensembl_species_history
status = mapping.status.id
sequence = None
if fetch_sequence:
try:
sequence = ensembl_sequence(mapping.transcript.enst_id,
ensembl_history.ensembl_release)
except Exception as e:
print(e) # TODO: log
sequence = None
mapping_obj = {
'mappingId': mapping.mapping_id,
'timeMapped': release_mapping_history.time_mapped,
'ensemblRelease': ensembl_history.ensembl_release,
'uniprotRelease': release_mapping_history.uniprot_release,
'uniprotEntry': {
'uniprotAccession': mapping.uniprot.uniprot_acc,
'entryType': Mapping.entry_type(mapping_history.entry_type_id),
'sequenceVersion': mapping.uniprot.sequence_version,
'upi': mapping.uniprot.upi,
'md5': mapping.uniprot.md5,
'isCanonical':
True if mapping.uniprot.canonical_uniprot_id else False,
'alias': mapping.uniprot.alias,
'ensemblDerived': mapping.uniprot.ensembl_derived,
'gene_symbol': mapping.uniprot.gene_symbol,
'gene_accession': mapping.uniprot.gene_accession,
'length': mapping.uniprot.length
},
'ensemblTranscript': {
'enstId': mapping.transcript.enst_id,
'enstVersion': mapping.transcript.enst_version,
'upi': mapping.transcript.uniparc_accession,
'biotype': mapping.transcript.biotype,
'deleted': mapping.transcript.deleted,
'chromosome': mapping.transcript.gene.chromosome,
'seqRegionStart': mapping.transcript.seq_region_start,
'seqRegionEnd': mapping.transcript.seq_region_end,
'seqRegionStrand': mapping.transcript.gene.seq_region_strand,
'ensgId': mapping.transcript.gene.ensg_id,
'ensgName': mapping.transcript.gene.gene_name,
'ensgSymbol': mapping.transcript.gene.gene_symbol,
'ensgAccession': mapping.transcript.gene.gene_accession,
'sequence': sequence,
'enspId': mapping.transcript.ensp_id,
'enspLen': mapping.transcript.ensp_len,
'select': mapping.transcript.select
},
'alignment_difference': mapping.alignment_difference,
'status': Mapping.status_type(status),
'status_history': mapping.statuses(usernames=authenticated)
}
return mapping_obj
def build_related_unmapped_entries_data(mapping):
"""
Return the list of unmapped entries releated to the mapping (via grouping_id)
"""
# related unmapped entries share the same grouping_id and tax id
mapping_mh = mapping.mapping_history.latest(
'release_mapping_history__time_mapped')
mapping_mh_rmh = mapping_mh.release_mapping_history
mapping_grouping_id = mapping_mh.grouping_id
related_unmapped_ue_histories = UniprotEntryHistory.objects.filter(
release_version=mapping_mh_rmh.uniprot_release,
grouping_id=mapping_grouping_id)
related_unmapped_ue_entries = []
for ueh in related_unmapped_ue_histories:
up_entry = ueh.uniprot
related_unmapped_ue_entries.append({
'uniprot_id':
up_entry.uniprot_id,
'uniprotAccession':
up_entry.uniprot_acc,
'entryType':
Mapping.entry_type(up_entry.entry_type_id),
'sequenceVersion':
up_entry.sequence_version,
'upi':
up_entry.upi,
'md5':
up_entry.md5,
'isCanonical':
not up_entry.canonical_uniprot_id,
'alias':
up_entry.alias,
'ensemblDerived':
up_entry.ensembl_derived,
'gene_symbol':
up_entry.gene_symbol,
'gene_accession':
up_entry.chromosome_line,
'length':
up_entry.length,
'protein_existence_id':
up_entry.protein_existence_id
})
related_unmapped_transcript_histories = TranscriptHistory.objects.filter(
ensembl_species_history=mapping_mh_rmh.ensembl_species_history,
grouping_id=mapping_grouping_id)
related_unmapped_transcripts = []
for t_hist in related_unmapped_transcript_histories:
transcript = t_hist.transcript
related_unmapped_transcripts.append({
'transcript_id':
transcript.transcript_id,
'enstId':
transcript.enst_id,
'enstVersion':
transcript.enst_version,
'upi':
transcript.uniparc_accession,
'biotype':
transcript.biotype,
'deleted':
transcript.deleted,
'chromosome':
transcript.gene.chromosome,
'regionAccession':
transcript.gene.region_accession,
'seqRegionStart':
transcript.seq_region_start,
'seqRegionEnd':
transcript.seq_region_end,
'seqRegionStrand':
transcript.gene.seq_region_strand,
'ensgId':
transcript.gene.ensg_id,
'ensgName':
transcript.gene.gene_name,
'ensgSymbol':
transcript.gene.gene_symbol,
'ensgAccession':
transcript.gene.gene_accession,
'ensgRegionAccession':
transcript.gene.region_accession,
'sequence':
None,
'enspId':
transcript.ensp_id,
'enspLen':
transcript.ensp_len,
'source':
transcript.source,
'select':
transcript.select
})
return {
'ensembl': related_unmapped_transcripts,
'uniprot': related_unmapped_ue_entries
}