Merge pull request #3366 from broadinstitute/shifa/variant-rnaseq-label

Add the RnaSeq Splice Junction outlier label to the variants.

seqr/fixtures/1kg_project.json

@@ -919,8 +919,8 @@
         "sample_id": "NA19679_S",
         "is_active": true,
         "elasticsearch_index":null,
-        "tissue_type": "M",
-        "data_source": "muscle_samples.tsv.gz",
+        "tissue_type": "F",
+        "data_source": "fibs_samples.tsv.gz",
         "loaded_date": "2017-02-05T06:42:55.397Z"
     }
 },
@@ -1376,7 +1376,7 @@
     "fields": {
         "sample": 151,
         "rank": 0,
-        "gene_id": "ENSG00000106554",
+        "gene_id": "ENSG00000268903",
         "p_value": 1.08E-56,
         "z_score": 12.34,
         "chrom": "7",
@@ -1416,7 +1416,7 @@
     "fields": {
         "sample": 151,
         "rank": 3,
-        "gene_id": "ENSG00000106554",
+        "gene_id": "ENSG00000268903",
         "p_value": 0.1,
         "z_score": 12.34,
         "chrom": "7",
@@ -1436,7 +1436,7 @@
     "fields": {
         "sample": 151,
         "rank": 1,
-        "gene_id": "ENSG00000106554",
+        "gene_id": "ENSG00000268903",
         "p_value": 0.0001,
         "z_score": 12.34,
         "chrom": "7",
@@ -1456,7 +1456,7 @@
     "fields": {
         "sample": 151,
         "rank": 2,
-        "gene_id": "ENSG00000106554",
+        "gene_id": "ENSG00000268903",
         "p_value": 0.0001,
         "z_score": 12.34,
         "chrom": "7",
@@ -1475,7 +1475,7 @@
     "pk": 6,
     "fields": {
         "sample": 152,
-        "gene_id": "ENSG00000106554",
+        "gene_id": "ENSG00000268903",
         "rank": 0,
         "p_value": 0.001,
         "z_score": 12.34,
@@ -1496,7 +1496,7 @@
     "fields": {
         "sample": 152,
         "rank": 1,
-        "gene_id": "ENSG00000106554",
+        "gene_id": "ENSG00000268903",
         "p_value": 0.2,
         "z_score": 12.34,
         "chrom": "7",

seqr/views/apis/individual_api_tests.py

@@ -999,17 +999,17 @@ def test_get_individual_rna_seq_data(self):
                 },
             },
             'spliceOutliers': {
-                'ENSG00000106554': mock.ANY,
+                'ENSG00000268903': mock.ANY,
             },
         }})
         self.assertDictEqual(
             {
-                'chrom': '7', 'deltaPsi': 0.85, 'end': 132886973, 'geneId': 'ENSG00000106554', 'isSignificant': True,
+                'chrom': '7', 'deltaPsi': 0.85, 'end': 132886973, 'geneId': 'ENSG00000268903', 'isSignificant': True,
                 'pValue': 1.08e-56, 'rareDiseaseSamplesTotal': 20, 'rareDiseaseSamplesWithJunction': 1,
                 'readCount': 1297, 'start': 132885746, 'strand': '*', 'type': 'psi5', 'zScore': 12.34,
                 'tissueType': 'F',
             },
-            response_json['rnaSeqData'][INDIVIDUAL_GUID]['spliceOutliers']['ENSG00000106554'][0]
+            response_json['rnaSeqData'][INDIVIDUAL_GUID]['spliceOutliers']['ENSG00000268903'][0]
         )
         self.assertSetEqual(set(response_json['genesById'].keys()), {'ENSG00000135953', 'ENSG00000268903'})
 
@@ -1026,7 +1026,7 @@ def test_get_individual_rna_seq_data_is_significant(self):
         self.assertEqual(2, len(significant_outliers))
         self.assertListEqual(
             [{field: outlier[field] for field in ['start', 'end', 'pValue', 'tissueType', 'isSignificant']}
-             for outlier in response_rnaseq_data['spliceOutliers']['ENSG00000106554']],
+             for outlier in response_rnaseq_data['spliceOutliers']['ENSG00000268903']],
             [{'start': 132885746, 'end': 132886973, 'pValue': 1.08e-56, 'tissueType': 'F', 'isSignificant': True},
              {'start': 1001, 'end': 2001, 'pValue': 0.1, 'tissueType': 'F', 'isSignificant': False},
              {'start': 3000, 'end': 4000, 'pValue': 0.0001, 'tissueType': 'F', 'isSignificant': True},

seqr/views/apis/saved_variant_api_tests.py

@@ -172,8 +172,10 @@ def test_saved_variant_data(self):
             'outliers': {
                 'ENSG00000135953': {
                     'geneId': 'ENSG00000135953', 'zScore': 7.31, 'pValue': 0.00000000000948, 'pAdjust': 0.00000000781,
-                    'isSignificant': True,
-            }},
+                    'tissueType': None, 'isSignificant': True,
+                }
+            },
+            'spliceOutliers': {},
         }})
 
         # include project tag types

seqr/views/apis/variant_search_api_tests.py

@@ -99,7 +99,10 @@
         'VFD0000025_1248367227_r0390_10': expected_functional_tag, 'VFD0000026_1248367227_r0390_10': expected_functional_tag,
     },
     'locusListsByGuid': {LOCUS_LIST_GUID: {'intervals': mock.ANY}},
-    'rnaSeqData': {'I000001_na19675': {'outliers': {'ENSG00000268903': mock.ANY}}},
+    'rnaSeqData': {
+        'I000001_na19675': {'outliers': {'ENSG00000268903': mock.ANY}, 'spliceOutliers': {'ENSG00000268903': mock.ANY}},
+        'I000003_na19679': {'outliers': {}, 'spliceOutliers': {'ENSG00000268903': mock.ANY}},
+    },
     'phenotypeGeneScores': {
         'I000001_na19675': {'ENSG00000268903': {'exomiser': EXPECTED_EXOMISER_DATA}},
         'I000002_na19678': {'ENSG00000268903': {'lirical': EXPECTED_LIRICAL_DATA}},
@@ -142,6 +145,7 @@ def _get_compound_het_es_variants(results_model, **kwargs):
     return deepcopy(COMP_HET_VARAINTS), 1
 
 
+@mock.patch('seqr.views.utils.orm_to_json_utils.RnaSeqSpliceOutlier.MAX_SIGNIFICANT_OUTLIER_NUM', 2)
 @mock.patch('seqr.views.utils.permissions_utils.safe_redis_get_json', lambda *args: None)
 class VariantSearchAPITest(object):
 
@@ -164,6 +168,24 @@ def _assert_expected_search_context(self, response_json):
         self.assertEqual(response_json['familiesByGuid']['F000001_1']['displayName'], '1')
         self.assertEqual(response_json['familiesByGuid']['F000001_1']['analysisStatus'], 'Q')
 
+    def _assert_expected_rnaseq_response(self, response_json):
+        self.assertDictEqual(
+            response_json['rnaSeqData']['I000001_na19675']['outliers']['ENSG00000268903'],
+            {'geneId': 'ENSG00000268903', 'isSignificant': True, 'pAdjust': 1.39e-09, 'pValue': 5.88e-10,
+             'tissueType': None, 'zScore': 7.08}
+        )
+        self.assertListEqual(
+            sorted(response_json['rnaSeqData']['I000001_na19675']['spliceOutliers']['ENSG00000268903'], key=lambda d: d['start']),
+            [{'chrom': '7', 'deltaPsi': 0.85, 'end': 4000, 'geneId': 'ENSG00000268903', 'isSignificant': True,
+              'pValue': 0.0001, 'rareDiseaseSamplesTotal': 20, 'rareDiseaseSamplesWithJunction': 1, 'readCount': 1297,
+              'start': 3000, 'strand': '*', 'tissueType': 'F', 'type': 'psi5', 'zScore': 12.34},
+             {'chrom': '7', 'deltaPsi': 0.85, 'end': 8000, 'geneId': 'ENSG00000268903', 'isSignificant': True,
+              'pValue': 0.001, 'rareDiseaseSamplesTotal': 20, 'rareDiseaseSamplesWithJunction': 1, 'readCount': 1297,
+              'start': 7000, 'strand': '*', 'tissueType': 'M', 'type': 'psi5', 'zScore': 12.34},
+             {'chrom': '7', 'deltaPsi': 0.85, 'end': 132886973, 'geneId': 'ENSG00000268903', 'isSignificant': True,
+              'pValue': 1.08e-56, 'rareDiseaseSamplesTotal': 20, 'rareDiseaseSamplesWithJunction': 1, 'readCount': 1297,
+              'start': 132885746, 'strand': '*', 'tissueType': 'F', 'type': 'psi5', 'zScore': 12.34}]
+        )
 
     def _assert_expected_results_family_context(self, response_json, locus_list_detail=False):
         self._assert_expected_results_context(response_json, locus_list_detail=locus_list_detail)
@@ -188,7 +210,9 @@ def _assert_expected_results_family_context(self, response_json, locus_list_deta
         self.assertEqual(len(response_json['familyNotesByGuid']), 3)
         self.assertSetEqual(set(response_json['familyNotesByGuid']['FAN000001_1'].keys()), FAMILY_NOTE_FIELDS)
 
-    def _assert_expected_results_context(self, response_json, has_pa_detail=True, locus_list_detail=False):
+        self._assert_expected_rnaseq_response(response_json)
+
+    def _assert_expected_results_context(self, response_json, has_pa_detail=True, locus_list_detail=False, rnaseq=True):
         gene_fields = {'locusListGuids'}
         gene_fields.update(GENE_VARIANT_FIELDS)
         basic_gene_id = next(gene_id for gene_id in ['ENSG00000268903', 'ENSG00000233653'] if gene_id in response_json['genesById'])
@@ -222,6 +246,8 @@ def _assert_expected_results_context(self, response_json, has_pa_detail=True, lo
         if response_json['variantFunctionalDataByGuid']:
             self.assertSetEqual(set(next(iter(response_json['variantFunctionalDataByGuid'].values())).keys()), FUNCTIONAL_FIELDS)
 
+        if rnaseq:
+            self._assert_expected_rnaseq_response(response_json)
 
     @mock.patch('seqr.utils.middleware.logger.error')
     @mock.patch('seqr.views.apis.variant_search_api.get_variant_query_gene_counts')
@@ -424,7 +450,7 @@ def test_query_variants(self, mock_get_variants, mock_get_gene_counts, mock_erro
         })
         expected_search_response['search']['totalResults'] = 1
         self.assertDictEqual(response_json, expected_search_response)
-        self._assert_expected_results_context(response_json, has_pa_detail=False)
+        self._assert_expected_results_context(response_json, has_pa_detail=False, rnaseq=False)
         mock_error_logger.assert_not_called()
 
         # Test cross-project discovery for analyst users
@@ -507,7 +533,6 @@ def _get_variants(results_model, **kwargs):
         response = self.client.post(url, content_type='application/json', data=json.dumps(body))
         self.assertEqual(response.status_code, 200)
         response_json = response.json()
-        self.assertSetEqual(set(response_json.keys()), set(EXPECTED_SEARCH_RESPONSE.keys()))
         self.assertDictEqual(response_json, EXPECTED_SEARCH_RESPONSE)
         self._assert_expected_results_context(response_json)
         self.assertSetEqual(

seqr/views/utils/variant_utils.py

@@ -1,13 +1,13 @@
 from collections import defaultdict
 from django.contrib.postgres.aggregates import ArrayAgg
-from django.db.models import F, Value
+from django.db.models import F
 import logging
 import redis
 
 from matchmaker.models import MatchmakerSubmissionGenes, MatchmakerSubmission
 from reference_data.models import TranscriptInfo
 from seqr.models import SavedVariant, VariantSearchResults, Family, LocusList, LocusListInterval, LocusListGene, \
-    RnaSeqOutlier, RnaSeqTpm, PhenotypePrioritization, Project
+    RnaSeqTpm, PhenotypePrioritization, Project
 from seqr.utils.search.utils import get_variants_for_variant_ids
 from seqr.utils.gene_utils import get_genes_for_variants
 from seqr.views.utils.json_to_orm_utils import update_model_from_json
@@ -131,19 +131,9 @@ def _add_locus_lists(projects, genes, add_list_detail=False, user=None):
 
 
 def _get_rna_seq_outliers(gene_ids, family_guids):
-    # TODO change to get_json_for_rna_seq_outliers in issue #3324
-    data_by_individual_gene = defaultdict(lambda: {'outliers': {}})
-
-    outlier_data = get_json_for_queryset(
-        RnaSeqOutlier.objects.filter(
-            gene_id__in=gene_ids, p_adjust__lt=RnaSeqOutlier.SIGNIFICANCE_THRESHOLD, sample__individual__family__guid__in=family_guids),
-        nested_fields=[{'fields': ('sample', 'individual', 'guid'), 'key': 'individualGuid'}],
-        additional_values={'isSignificant': Value(True)},
-    )
-    for data in outlier_data:
-        data_by_individual_gene[data.pop('individualGuid')]['outliers'][data['geneId']] = data
+    filters = {'gene_id__in': gene_ids, 'sample__individual__family__guid__in': family_guids}
 
-    return data_by_individual_gene
+    return get_json_for_rna_seq_outliers(filters)
 
 
 def get_phenotype_prioritization(family_guids, gene_ids=None):

ui/pages/Project/reducers.js

@@ -301,9 +301,11 @@ export const searchMmeMatches = submissionGuid => (dispatch) => {
 }
 
 export const loadRnaSeqData = individualGuid => (dispatch, getState) => {
-  const data = getState().rnaSeqDataByIndividual[individualGuid]
-  // If variants were loaded for the individual, the significant gene data will be loaded but not all the needed data
-  if (!data?.outliers || Object.values(data.outliers).every(({ isSignificant }) => isSignificant)) {
+  const { outliers, spliceOutliers } = getState().rnaSeqDataByIndividual[individualGuid] || {}
+  // If variants were loaded for the individual, the significant data were loaded but not the non-significant ones
+  if (!outliers || !spliceOutliers || (Object.values(outliers).every(({ isSignificant }) => isSignificant) &&
+    Object.values(spliceOutliers).flat().every(({ isSignificant }) => isSignificant))
+  ) {
     dispatch({ type: REQUEST_RNA_SEQ_DATA })
     new HttpRequestHelper(`/api/individual/${individualGuid}/rna_seq_data`,
       (responseJson) => {

ui/pages/Search/fixtures.js

@@ -18,7 +18,11 @@ export const LOCUS_LIST = {
   locusListGuid: LOCUS_LIST_GUID,
   name: "2017 Monogenic IBD Gene List",
   numEntries: 60,
-  parsedItems: { items:  [{ geneId: 'ENSG00000164458' }], itemMap: { 'TTN': { geneId: 'ENSG00000164458', symbol: 'TTN' } } }
+  parsedItems: { items:  [{ geneId: 'ENSG00000164458' }], itemMap: { 'TTN': { geneId: 'ENSG00000164458', symbol: 'TTN' } } },
+  intervals: [
+    {'chrom': '1', 'end': 7300, 'genomeVersion': '37', 'locusListGuid': 'LL00132_2017_monogenic_ibd_gen', 'locusListIntervalGuid': 'LLI0000012_test_list_edit4545_', 'start': 7200},
+    {'chrom': '3', 'end': 3000, 'genomeVersion': '37', 'locusListGuid': 'LL00132_2017_monogenic_ibd_gen', 'locusListIntervalGuid': 'LLI0000013_a_new_list325_3000', 'start': 25},
+  ]
 }
 
 export const STATE = {
@@ -100,6 +104,7 @@ export const STATE = {
       ],
       sampleGuids: [],
       sex: 'F',
+      familyGuid: FAMILY_GUID,
     },
     I021475_na19675: {
       affected: 'A',
@@ -212,6 +217,13 @@ export const STATE = {
     },
   },
   locusListsByGuid: { [LOCUS_LIST_GUID]: LOCUS_LIST },
+  rnaSeqDataByIndividual: { I021474_na19679: {
+    outliers: {ENSG00000136758: {geneId:  "ENSG00000136758", isSignificant: true, pAdjust: 0.000225907356686287, pValue: 2.69828505929319e-9, tissueType: "M", zScore: 5.62}},
+    spliceOutliers: { ENSG00000136758: [
+      {chrom: "10", deltaPsi: 0.56, end: 27114400, geneId: "ENSG00000136758", isSignificant: true, pValue: 2.1234e-10, rareDiseaseSamplesTotal: 171, rareDiseaseSamplesWithJunction: 1, readCount: 1208, start: 27114300, strand: "*", tissueType: "F", type: "psi5", zScore: 2.96},
+      {chrom: "11", deltaPsi: 0.56, end: 27114400, geneId: "ENSG00000136758", isSignificant: true, pValue: 2.1234e-10, rareDiseaseSamplesTotal: 171, rareDiseaseSamplesWithJunction: 1, readCount: 1208, start: 27114300, strand: "*", tissueType: "F", type: "psi5", zScore: 2.96},
+    ]},
+  }},
   projectsByGuid: {
     [PROJECT_GUID] : {
       createdDate: '2016-05-16T05:37:08.634Z',

ui/redux/selectors.js

@@ -360,6 +360,20 @@ export const getSearchGeneBreakdownValues = createSelector(
   })),
 )
 
+const groupDataNestedByChrom = (initialData, groupedData, nestedKey) => groupedData.reduce(
+  (acc, data) => {
+    const { chrom } = data
+    if (!acc[chrom]) {
+      acc[chrom] = {}
+    }
+    if (!acc[chrom][nestedKey]) {
+      acc[chrom][nestedKey] = []
+    }
+    acc[chrom][nestedKey].push(data)
+    return acc
+  }, initialData,
+)
+
 export const getLocusListIntervalsByChromProject = createSelector(
   getProjectsByGuid,
   getLocusListsByGuid,
@@ -368,16 +382,7 @@ export const getLocusListIntervalsByChromProject = createSelector(
       const projectIntervals = locusListGuids.map(locusListGuid => locusListsByGuid[locusListGuid]).reduce(
         (acc2, { intervals = [] }) => [...acc2, ...intervals], [],
       )
-      projectIntervals.forEach((interval) => {
-        if (!acc[interval.chrom]) {
-          acc[interval.chrom] = {}
-        }
-        if (!acc[interval.chrom][projectGuid]) {
-          acc[interval.chrom][projectGuid] = []
-        }
-        acc[interval.chrom][projectGuid].push(interval)
-      })
-      return acc
+      return groupDataNestedByChrom(acc, projectIntervals, projectGuid)
     }, {},
   ),
 )
@@ -419,18 +424,27 @@ export const getRnaSeqSignificantJunctionData = createSelector(
   getGenesById,
   getIndividualsByGuid,
   getRnaSeqDataByIndividual,
-  (genesById, individualsByGuid, rnaSeqDataByIndividual) => Object.entries(rnaSeqDataByIndividual).reduce(
-    (acc, [individualGuid, rnaSeqData]) => (rnaSeqData.spliceOutliers ? {
-      ...acc,
-      [individualGuid]: Object.values(rnaSeqData.spliceOutliers).flat().filter(({ isSignificant }) => isSignificant)
+  (genesById, individualsByGuid, rnaSeqDataByIndividual) => Object.entries(rnaSeqDataByIndividual || {}).reduce(
+    (acc, [individualGuid, rnaSeqData]) => {
+      const individualData = Object.values(rnaSeqData.spliceOutliers || {}).flat()
+        .filter(({ isSignificant }) => isSignificant)
         .sort((a, b) => a.pValue - b.pValue)
         .map(({ geneId, chrom, start, end, strand, type, ...cols }) => ({
           geneSymbol: (genesById[geneId] || {}).geneSymbol || geneId,
           idField: `${geneId}-${chrom}-${start}-${end}-${strand}-${type}`,
           familyGuid: individualsByGuid[individualGuid].familyGuid,
+          individualName: individualsByGuid[individualGuid].displayName,
           individualGuid,
           ...{ geneId, chrom, start, end, strand, type, ...cols },
-        })),
-    } : acc), {},
+        }))
+      return (individualData.length > 0 ? { ...acc, [individualGuid]: individualData } : acc)
+    }, {},
+  ),
+)
+
+export const getSpliceOutliersByChromFamily = createSelector(
+  getRnaSeqSignificantJunctionData,
+  spliceDataByIndiv => Object.values(spliceDataByIndiv).reduce(
+    (acc, spliceData) => (groupDataNestedByChrom(acc, spliceData, spliceData[0].familyGuid)), {},
   ),
 )

ui/redux/selectors.test.js

@@ -6,6 +6,8 @@ import {
   getSearchGeneBreakdownValues,
   getSelectableTagTypesByProject,
   getUserOptions,
+  getLocusListIntervalsByChromProject,
+  getSpliceOutliersByChromFamily,
 } from './selectors'
 import {FAMILY_GUID, GENE_ID, SEARCH, SEARCH_HASH, STATE} from "../pages/Search/fixtures";
 
@@ -47,3 +49,39 @@ test('getUserOptions', () => {
   expect(Object.keys(options).length).toEqual(7)
   expect(options[1]).toEqual({ key: '4MW8vPtmHG', value: '4MW8vPtmHG', text: 'Mekdes ([email protected])'})
 })
+
+test('getLocusListIntervalsByChromProject', () => {
+  expect(getLocusListIntervalsByChromProject(STATE, {})).toEqual({
+    ['1']: {
+      'R0237_1000_genomes_demo': [
+        {'chrom': '1', 'end': 7300, 'genomeVersion': '37', 'locusListGuid': 'LL00132_2017_monogenic_ibd_gen', 'locusListIntervalGuid': 'LLI0000012_test_list_edit4545_', 'start': 7200},
+      ],
+    },
+    ['3']: {
+      'R0237_1000_genomes_demo': [
+        {'chrom': '3', 'end': 3000, 'genomeVersion': '37', 'locusListGuid': 'LL00132_2017_monogenic_ibd_gen', 'locusListIntervalGuid': 'LLI0000013_a_new_list325_3000', 'start': 25},
+      ],
+    }
+  })
+})
+
+test('getSpliceOutliersByChromFamily', () => {
+  expect(getSpliceOutliersByChromFamily(STATE, {})).toEqual({
+    ['10']: {
+      'F011652_1': [
+        {
+          familyGuid: "F011652_1", geneSymbol: "ENSG00000136758", idField: "ENSG00000136758-10-27114300-27114400-*-psi5", individualGuid: "I021474_na19679", individualName: "",
+          chrom: "10", deltaPsi: 0.56, end: 27114400, geneId: "ENSG00000136758", isSignificant: true, pValue: 2.1234e-10, rareDiseaseSamplesTotal: 171, rareDiseaseSamplesWithJunction: 1, readCount: 1208, start: 27114300, strand: "*", tissueType: "F", type: "psi5", zScore: 2.96,
+        },
+      ],
+    },
+    ['11']: {
+      'F011652_1': [
+        {
+          familyGuid: "F011652_1", geneSymbol: "ENSG00000136758", idField: "ENSG00000136758-11-27114300-27114400-*-psi5", individualGuid: "I021474_na19679", individualName: "",
+          chrom: "11", deltaPsi: 0.56, end: 27114400, geneId: "ENSG00000136758", isSignificant: true, pValue: 2.1234e-10, rareDiseaseSamplesTotal: 171, rareDiseaseSamplesWithJunction: 1, readCount: 1208, start: 27114300, strand: "*", tissueType: "F", type: "psi5", zScore: 2.96,
+        },
+      ],
+    }
+  })
+})