C4-294 Named Genotype Label Fields (#200)

* C4-196 intermediate update * Schema and types file * test inserts * add demo_inserts * less gene lists for demo * update uuid * update mapping * add default filter set for case * add gene list digester * update delimiters and attachment type * add ensemble look up * add xls xlsx * small changes in support of moving to es6 * add poetry.lock back * C4-196 ES6 compatible cgap * C4-294 add sample_id to genotype labels * C4-294 clean up, spec out calc prop * C4-294 first pass, always proband centric view * add omitted items() call * add tests * update versions and add pandas * add more tests * Commented out test for genelist processing * default values in calc prop * rip out es6 compat changes from this branch * add revlink and test * add demo insert * remove duplicates * fix type * remove ingester * C4-294 mapping table update + genotype label facets * remove pandas * resolve lock file * C4-294 scrub label * fix cmphet, fix test Co-authored-by: Koray Kırlı <koray_kirli@hms.harvard.edu>
dbmi-bgm · Sep 4, 2020 · f352425 · f352425
1 parent 02e10d8
commit f352425
Show file tree

Hide file tree

Showing 29 changed files with 1,295 additions and 882 deletions.
diff --git a/poetry.lock b/poetry.lock
diff --git a/pyproject.toml b/pyproject.toml
@@ -1,7 +1,7 @@
 [tool.poetry]
 # Note: Various modules refer to this system as "encoded", not "cgap-portal".
 name = "encoded"
-version = "2.4.20"
+version = "2.5.0"
 description = "Clinical Genomics Analysis Platform"
 authors = ["4DN-DCIC Team <support@4dnucleome.org>"]
 license = "MIT"
@@ -35,7 +35,7 @@ classifiers = [
 ]
 
 [tool.poetry.dependencies]
-python = ">=3.6,<3.7"
+python = ">=3.6.1,<3.7"
 boto3 = "^1.10.46"
 botocore = "^1.13.46"
 certifi = ">=2020.4.5.2"
@@ -80,7 +80,7 @@ pyramid-multiauth = "0.8.0"
 pyramid-retry = "1.0"
 pyramid-tm = "2.2.1"
 pyramid-translogger = "^0.1"
-python-dateutil = "2.5.3"
+python-dateutil = "2.7.3"
 # python-magic is presently pinned to 0.4.15 in lockstep with dcicsnovault's requirements. See explanation there.
 python-magic = "0.4.15"
 pytz = ">=2020.1"

diff --git a/src/encoded/annotations/variant_table_v0.4.8.csv b/src/encoded/annotations/variant_table_v0.4.8.csv
diff --git a/src/encoded/commands/create_mapping_on_deploy.py b/src/encoded/commands/create_mapping_on_deploy.py
@@ -33,6 +33,7 @@
     'FileReference',
     'Image',
     'Gene',
+    'GeneList',
     'Phenotype',
     'Disorder',
     'Individual',

diff --git a/src/encoded/commands/ingest_vcf.py b/src/encoded/commands/ingest_vcf.py
@@ -547,7 +547,10 @@ def create_sample_variant_from_record(self, record):
                         for group in comhet:
                             annotations = {}
                             for field_name, value in zip(field_names, group.split('|')):
-                                annotations[field_name] = self.fix_encoding(value)
+                                if field_name == 'comhet_transcript':  # array field
+                                    annotations[field_name] = self.fix_encoding(value).split('~')
+                                else:
+                                    annotations[field_name] = self.fix_encoding(value)
                             s['cmphet'].append(annotations)
 
             self.parse_samples(s, sample)  # add sample fields, already formatted

diff --git a/src/encoded/commands/variant_table_intake.py b/src/encoded/commands/variant_table_intake.py
@@ -457,6 +457,29 @@ def add_extra_variant_sample_columns(cols):
             'title': 'Variant'
         }
 
+    @staticmethod
+    def add_extra_variant_sample_facets(facs):
+        facs['associated_genotype_labels.proband_genotype_label'] = {
+            'title': 'Proband Genotype',
+            'order': 12,
+            'grouping': 'Genotype'
+        }
+        facs['associated_genotype_labels.mother_genotype_label'] = {
+            'title': 'Mother Genotype',
+            'order': 13,
+            'grouping': 'Genotype'
+        }
+        facs['associated_genotype_labels.father_genotype_label'] = {
+            'title': 'Father Genotype',
+            'order': 14,
+            'grouping': 'Genotype'
+        }
+        facs['inheritance_modes'] = {
+            'title': 'Inheritance Modes',
+            'order': 15,
+            'grouping': 'Genotype'
+        }
+
     def generate_variant_sample_schema(self, sample_props, cols, facs, variant_cols, variant_facs):
         """ Builds the variant_sample.json schema based on sample_props. Will also add variant columns and
             facets since this information is embedded
@@ -548,6 +571,7 @@ def format_variant_cols_or_facs(d):
         cols.update(variant_cols)  # add variant stuff since we are embedding this info
         facs.update(variant_facs)
         self.add_extra_variant_sample_columns(cols)
+        self.add_extra_variant_sample_facets(facs)
         schema['columns'] = cols
         schema['facets'] = facs
         schema['facets'] = self.sort_schema_properties(schema, key='facets')

diff --git a/src/encoded/inheritance_mode.py b/src/encoded/inheritance_mode.py
@@ -11,7 +11,7 @@ class InheritanceModeError(Exception):
 
 class InheritanceMode:
 
-    MISSING = '.'  # XXX: is this really what this is? Should it be called 'dot'?
+    MISSING = '.'
 
     AUTOSOME = 'autosome'
     CHROMOSOMES = [
@@ -25,7 +25,7 @@ class InheritanceMode:
 
     MOTHER = 'mother'
     FATHER = 'father'
-    SELF = 'proband'  # XXX: this is what the data actually looks like, NOT 'self'
+    SELF = 'proband'
     TRIO = [MOTHER, FATHER, SELF]
 
     # Genotype labels
@@ -308,7 +308,7 @@ def compute_cmphet_inheritance_modes(cmphet):
         return inheritance_modes
 
     @staticmethod
-    def build_genotype_label_structure(genotype_labels):
+    def build_genotype_label_structure(genotype_labels, sample_ids):
         """ Converts the genotype_labels structure into a consistent structure that can be used
             in our item ecosystem.
 
@@ -323,7 +323,8 @@ def build_genotype_label_structure(genotype_labels):
         for role, labels in genotype_labels.items():
             structured_labels.append({
                 'role': role,
-                'labels': labels
+                'labels': labels,
+                'sample_id': sample_ids.get(role)
             })
         return structured_labels
 
@@ -338,6 +339,7 @@ def compute_inheritance_modes(cls, variant_sample, chrom=None):
         """
         sample_geno = variant_sample.get('samplegeno', [])
         try:
+            sample_ids = {s["samplegeno_role"]: s["samplegeno_sampleid"] for s in sample_geno}
             genotypes = {s["samplegeno_role"]: s["samplegeno_numgt"] for s in sample_geno}
             sexes = {s["samplegeno_role"]: s["samplegeno_sex"] for s in sample_geno}
             chrom = chrom if chrom else variant_sample.get('variant', {}).get('CHROM')  # attempt to get from variant
@@ -349,7 +351,7 @@ def compute_inheritance_modes(cls, variant_sample, chrom=None):
             return {}
 
         if chrom not in ['X', 'Y']:
-            chrom = cls.AUTOSOME  # XXX: so chrom is one of ['X', 'Y', 'autosome'] ?
+            chrom = cls.AUTOSOME
 
         if cls.SELF not in genotypes:
             raise InheritanceModeError('Role "proband" not present in genotypes: %s' % genotypes)
@@ -363,7 +365,7 @@ def compute_inheritance_modes(cls, variant_sample, chrom=None):
             inheritance_modes = cls.inheritance_modes_other_labels(genotypes, genotype_labels)
 
         new_fields = {
-            'genotype_labels': cls.build_genotype_label_structure(genotype_labels),
+            'genotype_labels': cls.build_genotype_label_structure(genotype_labels, sample_ids),
             'inheritance_modes': inheritance_modes
         }
 

diff --git a/src/encoded/schemas/gene_list.json b/src/encoded/schemas/gene_list.json
@@ -0,0 +1,59 @@
+{
+    "title": "Gene List",
+    "description": "Groups of genes that are relecant for a disease or condition",
+    "id": "/profiles/gene_list.json",
+    "$schema": "http://json-schema.org/draft-04/schema#",
+    "type": "object",
+    "required": ["institution", "project", "title"],
+    "identifyingProperties": ["uuid", "aliases"],
+    "additionalProperties": false,
+    "mixinProperties": [
+        { "$ref": "mixins.json#/schema_version" },
+        { "$ref": "mixins.json#/uuid" },
+        { "$ref": "mixins.json#/status" },
+        { "$ref": "mixins.json#/aliases" },
+        { "$ref": "mixins.json#/attribution" },
+        { "$ref": "mixins.json#/submitted" },
+        { "$ref": "mixins.json#/modified" },
+        { "$ref": "mixins.json#/tags" },
+        { "$ref": "mixins.json#/notes" }
+    ],
+    "properties": {
+        "schema_version": {
+            "default": "1"
+        },
+        "title": {
+            "title": "Title",
+            "description": "Title for this Gene List",
+            "uniqueItems": true,
+            "type": "string"
+        },
+        "genes" : {
+            "title" : "Genes",
+            "type" : "array",
+            "uniqueItems": true,
+            "items" : {
+                "title" : "Gene",
+                "type" : "string",
+                "linkTo" : "Gene"
+            }
+        },
+        "disorders" : {
+            "title" : "Related Disorders",
+            "description": "Disorders that are related to this gene list",
+            "type" : "array",
+            "uniqueItems": true,
+            "items" : {
+                "title" : "Disorder",
+                "type" : "string",
+                "linkTo" : "Disorder"
+            }
+        },
+        "source_file": {
+            "title" : "Source File",
+            "description": "The original gene list file used for generating this gene list item",
+            "type" : "string",
+            "linkTo": "Document"
+        }
+    }
+}