Create a Gene Shet reference data update cmd.

CHANGELOG.md

@@ -2,6 +2,7 @@
 
 ## dev
 * Migrate Fammily post_discovery_omim_number to integer array (REQUIRES DB MIGRATION)
+* Add GeneShet model to the reference DB (REQUIRES DB MIGRATION)
 
 ## 10/6/23
 * Require tissue_type in Sample model (REQUIRES DB MIGRATION)

reference_data/management/commands/update_gene_shet.py

@@ -0,0 +1,22 @@
+import logging
+from reference_data.management.commands.utils.update_utils import GeneCommand, ReferenceDataHandler
+from reference_data.models import GeneShet
+
+logger = logging.getLogger(__name__)
+
+
+class ShetReferenceDataHandler(ReferenceDataHandler):
+
+    model_cls = GeneShet
+    url = 'https://zenodo.org/record/7939768/files/s_het_estimates.genebayes.tsv'
+
+    @staticmethod
+    def parse_record(record):
+        yield {
+            'gene_id': record['ensg'],
+            'post_mean': float(record['post_mean']),
+        }
+
+
+class Command(GeneCommand):
+    reference_data_handler = ShetReferenceDataHandler

reference_data/management/tests/update_gene_shet_tests.py

@@ -0,0 +1,20 @@
+from reference_data.models import GeneShet
+from reference_data.management.tests.test_utils import ReferenceDataCommandTestCase
+
+class UpdateGeneShetTest(ReferenceDataCommandTestCase):
+    URL = 'https://zenodo.org/record/7939768/files/s_het_estimates.genebayes.tsv'
+    DATA = [
+        'ensg	hgnc	chrom	obs_lof	exp_lof	prior_mean	post_mean	post_lower_95	post_upper_95\n',
+        'ENSG00000223972	HGNC:37225	chr15	26.0	21.66	0.00059216	3.01e-05	1.05e-06	0.00010405\n',
+        'ENSG00000227233	HGNC:26441	chr5	31.0	28.55	0.00038727	4.853e-05	3.05e-06	0.00015705\n',
+        'ENSG00000243485	HGNC:4013	chr19	17.0	11.327	0.00082297	5.083e-05	3.05e-06	0.00016605\n'
+    ]
+
+    def test_update_gene_cn_sensitivity_command(self):
+        self._test_update_command('update_gene_shet', 'GeneShet', created_records=2)
+
+        self.assertEqual(GeneShet.objects.count(), 2)
+        record = GeneShet.objects.get(gene__gene_id='ENSG00000223972')
+        self.assertEqual(record.post_mean, 3.01E-05)
+        record = GeneShet.objects.get(gene__gene_id='ENSG00000243485')
+        self.assertEqual(record.post_mean, 5.083E-05)

reference_data/migrations/0022_geneshet.py

@@ -0,0 +1,22 @@
+# Generated by Django 3.2.18 on 2023-10-10 20:12
+
+from django.db import migrations, models
+import django.db.models.deletion
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ('reference_data', '0021_auto_20221031_2049'),
+    ]
+
+    operations = [
+        migrations.CreateModel(
+            name='GeneShet',
+            fields=[
+                ('id', models.AutoField(auto_created=True, primary_key=True, serialize=False, verbose_name='ID')),
+                ('post_mean', models.FloatField()),
+                ('gene', models.ForeignKey(on_delete=django.db.models.deletion.CASCADE, to='reference_data.geneinfo')),
+            ],
+        ),
+    ]

reference_data/models.py

@@ -161,6 +161,15 @@ class Meta:
         json_fields = ['pHI', 'pTS']
 
 
+class GeneShet(models.Model):
+    gene = models.ForeignKey(GeneInfo, on_delete=models.CASCADE)
+
+    post_mean = models.FloatField()
+
+    class Meta:
+        json_fields = ['post_mean']
+
+
 class Omim(models.Model):
     MAP_METHOD_CHOICES = (
         ('1', 'the disorder is placed on the map based on its association with a gene, but the underlying defect is not known.'),

seqr/fixtures/reference_data.json

@@ -1167,6 +1167,14 @@
         "pHI": 0.90576,
         "pTS": 0.7346
     }
+},
+{
+    "model": "reference_data.geneshet",
+    "pk": 1,
+    "fields": {
+        "gene": 1,
+        "post_mean": 0.90576
+    }
 },
     {
     "model": "reference_data.dbnsfpgene",

seqr/utils/gene_utils.py

@@ -4,7 +4,7 @@
 from django.db.models.functions import Length
 
 from reference_data.models import GeneInfo, GeneConstraint, dbNSFPGene, Omim, MGI, PrimateAI, GeneCopyNumberSensitivity, \
-    GenCC, ClinGen
+    GenCC, ClinGen, GeneShet
 from seqr.utils.xpos_utils import get_xpos
 from seqr.views.utils.orm_to_json_utils import _get_json_for_model, _get_json_for_models, _get_empty_json_for_model, \
     get_json_for_gene_notes_by_gene_id
@@ -90,6 +90,7 @@ def _add_mgi(gene):
 OMIM = 'omim'
 CONSTRAINT = 'constraint'
 CN_SENSITIVITY = 'cn_sensitivity'
+SHET = 'shet'
 DBNSFP = 'dbnsfp'
 GENCC = 'gencc'
 PRIMATE_AI = 'primate_ai'
@@ -100,6 +101,7 @@ def _add_mgi(gene):
     OMIM: (Omim, _add_omim),
     CONSTRAINT: (GeneConstraint, None),
     CN_SENSITIVITY: (GeneCopyNumberSensitivity, _add_gene_model('genecopynumbersensitivity', 'cnSensitivity', dict)),
+    SHET: (GeneShet, _add_gene_model('geneshet', 'sHet', dict)),
     GENCC: (GenCC, _add_gene_model('gencc', 'genCc', dict)),
     CLINGEN: (ClinGen, _add_gene_model('clingen', 'clinGen', lambda: None)),
 }

seqr/views/apis/family_api_tests.py

@@ -545,7 +545,7 @@ def test_get_family_phenotype_gene_scores(self):
                 'ENSG00000268903': {
                     'chromGrch37': '1', 'chromGrch38': '1', 'clinGen': None, 'cnSensitivity': {},
                     'codingRegionSizeGrch37': 0, 'codingRegionSizeGrch38': 0, 'constraints': {},
-                    'endGrch37': 135895, 'endGrch38': 135895, 'genCc': {},
+                    'endGrch37': 135895, 'endGrch38': 135895, 'genCc': {}, 'sHet': {},
                     'gencodeGeneType': 'processed_pseudogene', 'geneId': 'ENSG00000268903',
                     'geneSymbol': 'AL627309.7', 'mimNumber': None, 'omimPhenotypes': [],
                     'startGrch37': 135141, 'startGrch38': 135141

seqr/views/apis/gene_api_tests.py

@@ -33,6 +33,35 @@ def test_genes_info(self):
         genes = response.json()['genesById']
         self.assertSetEqual(set(genes.keys()), {GENE_ID, 'ENSG00000269981'})
         self.assertSetEqual(set(genes[GENE_ID].keys()), GENE_DETAIL_FIELDS)
+        self.assertDictEqual(genes[GENE_ID], {
+            'chromGrch37': '1',
+            'chromGrch38': '1',
+            'clinGen': {'haploinsufficiency': 'No Evidence', 'href': 'https://dosage.clinicalgenome.org/clingen_gene.cgi?sym=', 'triplosensitivity': ''},
+            'cnSensitivity': {'phi': 0.90576, 'pts': 0.7346},
+            'codingRegionSizeGrch37': 0,
+            'codingRegionSizeGrch38': 0,
+            'constraints': {'louef': 1.606, 'louefRank': 0, 'misZ': -0.7773, 'misZRank': 1, 'pli': 0.00090576, 'pliRank': 1, 'totalGenes': 1},
+            'diseaseDesc': '',
+            'endGrch37': 14409,
+            'endGrch38': 14409,
+            'functionDesc': '',
+            'genCc': {'hgncId': 'HGNC:943', 'classifications': [
+                {'classification': 'Strong', 'date': '7/29/19 19:04', 'disease': 'dystonia 16', 'moi': 'Autosomal recessive', 'submitter': 'Laboratory for Molecular Medicine'},
+                {'classification': 'Supportive', 'date': '9/14/21 0:00', 'disease': 'dystonia 16', 'moi': 'Autosomal recessive', 'submitter': 'Orphanet'},
+            ]},
+            'gencodeGeneType': 'transcribed_unprocessed_pseudogene',
+            'geneId': 'ENSG00000223972',
+            'geneNames': '',
+            'geneSymbol': 'DDX11L1',
+            'mgiMarkerId': None,
+            'mimNumber': 147571,
+            'notes': [],
+            'omimPhenotypes': [{'mimNumber': 147571, 'phenotypeDescription': 'Immunodeficiency 38', 'phenotypeInheritance': 'Autosomal recessive', 'phenotypeMimNumber': 616126}],
+            'primateAi': {'percentile25': 0.587214291096, 'percentile75': 0.821286439896},
+            'sHet': {'postMean': 0.90576},
+            'startGrch37': 11869,
+            'startGrch38': 11869,
+        })
 
 
     def test_create_update_and_delete_gene_note(self):

seqr/views/utils/test_utils.py

@@ -794,7 +794,7 @@ def _get_list_param(call, param):
     'gencodeGeneType', 'geneId', 'geneSymbol', 'startGrch37', 'startGrch38',
 }
 GENE_VARIANT_DISPLAY_FIELDS = {
-    'constraints', 'omimPhenotypes', 'mimNumber', 'cnSensitivity', 'genCc', 'clinGen',
+    'constraints', 'omimPhenotypes', 'mimNumber', 'cnSensitivity', 'genCc', 'clinGen', 'sHet',
 }
 GENE_VARIANT_DISPLAY_FIELDS.update(GENE_FIELDS)
 GENE_VARIANT_FIELDS = {

ui/shared/components/panel/genes/GeneDetail.jsx

@@ -175,6 +175,7 @@ DosageSensitivity.propTypes = {
 
 export const HI_THRESHOLD = 0.86
 export const TS_THRESHOLD = 0.94
+export const SHET_THRESHOLD = 0.1
 const HAPLOINSUFFICIENT_FIELDS = [{ field: 'phi', label: 'pHaplo' }]
 const TRIPLOSENSITIVE_FIELDS = [{ field: 'pts', label: 'pTriplo' }]
 const STAT_DETAILS = [
@@ -210,6 +211,24 @@ const STAT_DETAILS = [
     note: 'These metrics are based on the amount of expected variation observed in the gnomAD data and is a measure ' +
     'of how likely the gene is to be intolerant of loss-of-function mutations.',
   },
+  {
+    title: 'Shet',
+    scoreField: 'sHet',
+    fields: [
+      { field: 'postMean', label: 'post_mean' },
+    ],
+    note: (
+      <span>
+        This score was developed by the Pritchard lab [
+        <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10245655" target="_blank" rel="noreferrer">
+          Zeng et al 2023
+        </a>
+        ] to predict gene constraint based on functional and evolutionary information. Scores &gt;
+        {SHET_THRESHOLD}
+        &nbsp; are considered to have high likelihood to be under extreme selection.
+      </span>
+    ),
+  },
   {
     title: 'Haploinsufficient',
     content: gene => (

ui/shared/components/panel/variants/VariantGene.jsx

@@ -23,7 +23,7 @@ import { InlineHeader, NoBorderTable, ButtonLink, ColoredLabel } from '../../Sty
 import { PermissiveGeneSearchLink } from '../../buttons/SearchResultsLink'
 import ShowGeneModal from '../../buttons/ShowGeneModal'
 import Modal from '../../modal/Modal'
-import { GenCC, ClingenLabel, HI_THRESHOLD, TS_THRESHOLD } from '../genes/GeneDetail'
+import { GenCC, ClingenLabel, HI_THRESHOLD, TS_THRESHOLD, SHET_THRESHOLD } from '../genes/GeneDetail'
 import { getIndividualGeneDataByFamilyGene } from './selectors'
 
 const RnaSeqTpm = React.lazy(() => import('./RnaSeqTpm'))
@@ -379,24 +379,38 @@ const GENE_DETAIL_SECTIONS = [
     color: 'red',
     description: 'Loss of Function Constraint',
     label: 'LOF CONSTR',
-    showDetails: gene => gene.constraints.louef < LOF_THRESHHOLD,
+    showDetails: gene => (gene.constraints.louef < LOF_THRESHHOLD) ||
+      (gene.cnSensitivity.phi && gene.cnSensitivity.phi > HI_THRESHOLD) ||
+      (gene.sHet.postMean && gene.sHet.postMean > SHET_THRESHOLD),
     detailsDisplay: gene => (
-      `This gene ranks as ${gene.constraints.louefRank} most intolerant of LoF mutations out of
-       ${gene.constraints.totalGenes} genes under study (louef:
-       ${gene.constraints.louef.toPrecision(4)}${gene.constraints.pli ? `, pLi: ${gene.constraints.pli.toPrecision(4)}` : ''}).
-       LOEUF is the observed to expected upper bound fraction for loss-of-function variants based on the variation
-       observed in the gnomad data. Both LOEUF and pLi are measures of how likely the gene is to be intolerant of
-       loss-of-function mutations`),
-  },
-  {
-    color: 'red',
-    description: 'HaploInsufficient',
-    label: 'HI',
-    showDetails: gene => gene.cnSensitivity.phi && gene.cnSensitivity.phi > HI_THRESHOLD,
-    detailsDisplay: gene => (
-      `These are a score developed by the Talkowski lab that predict whether a gene is haploinsufficient based 
-      on large chromosomal microarray data set analysis. Scores >${HI_THRESHOLD} are considered to have high likelihood to be 
-      haploinsufficient. This gene has a score of ${gene.cnSensitivity.phi.toPrecision(4)}.`),
+      <List bulleted>
+        <List.Item>
+          This gene ranks as &nbsp;
+          {gene.constraints.louefRank}
+          &nbsp;most intolerant of LoF mutations out of &nbsp;
+          {gene.constraints.totalGenes}
+          &nbsp;genes under study (louef: &nbsp;
+          {gene.constraints.louef.toPrecision(4)}
+          {gene.constraints.pli ? `, pLi: ${gene.constraints.pli.toPrecision(4)}` : ''}
+          )
+          <a href="https://pubmed.ncbi.nlm.nih.gov/32461654/" target="_blank" rel="noreferrer"> Karczewski (2020)</a>
+        </List.Item>
+        {gene.sHet.postMean && (
+          <List.Item>
+            This gene has a Shet score of &nbsp;
+            {gene.sHet.postMean.toPrecision(4)}
+            <a href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10245655" target="_blank" rel="noreferrer"> Zeng (2023)</a>
+          </List.Item>
+        )}
+        {gene.cnSensitivity.phi && (
+          <List.Item>
+            This gene has a haploinsufficiency (HI) score of &nbsp;
+            {gene.cnSensitivity.phi.toPrecision(4)}
+            <a href="https://pubmed.ncbi.nlm.nih.gov/35917817" target="_blank" rel="noreferrer"> Collins (2022)</a>
+          </List.Item>
+        )}
+      </List>
+    ),
   },
   {
     color: 'red',