Merge pull request #81 from VariantEffect/bencap/feature/60/genomic-coordinate-score-sets

Add Support for Genomic Coordinate Score Sets

Author: bencap

alembic/versions/194cfebabe32_rename_wild_type_sequence.py

@@ -0,0 +1,32 @@
+"""Rename Wild Type Sequence
+
+Revision ID: 194cfebabe32
+Revises: 44d5c568f64b
+Create Date: 2023-08-29 12:48:18.390567
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "194cfebabe32"
+down_revision = "44d5c568f64b"
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    op.rename_table("wild_type_sequences", "target_sequences")
+    op.alter_column("target_genes", "wt_sequence_id", new_column_name="target_sequence_id")
+    op.execute("ALTER SEQUENCE  wild_type_sequences_id_seq RENAME TO target_sequences_id_seq")
+    op.execute("ALTER INDEX wild_type_sequences_pkey RENAME TO target_sequences_pkey")
+    op.execute("ALTER INDEX ix_wild_type_sequences_id RENAME TO ix_target_sequences_id")
+
+
+def downgrade():
+    op.rename_table("target_sequences", "wild_type_sequences")
+    op.alter_column("target_genes", "target_sequence_id", new_column_name="wt_sequence_id")
+    op.execute("ALTER SEQUENCE  target_sequences_id_seq RENAME TO wild_type_sequences_id_seq")
+    op.execute("ALTER INDEX target_sequences_pkey RENAME TO wild_type_sequences_pkey")
+    op.execute("ALTER INDEX ix_target_sequences_id RENAME TO ix_wild_type_sequences_id")

alembic/versions/44d5c568f64b_simplify_reference_genome_target_.py

@@ -0,0 +1,61 @@
+"""Simplify Reference Genome Target Structure
+
+Revision ID: 44d5c568f64b
+Revises: 90e7860964a2
+Create Date: 2023-08-24 15:20:01.208691
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "44d5c568f64b"
+down_revision = "90e7860964a2"
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    op.add_column(
+        "wild_type_sequences",
+        sa.Column("reference_id", sa.Integer, sa.ForeignKey("reference_genomes.id"), nullable=True),
+    )
+    op.execute(
+        """
+                UPDATE wild_type_sequences w
+                    SET reference_id = l.genome_id
+                    FROM
+                        (select * from reference_maps inner join target_genes on target_genes.id = reference_maps.target_id) as l
+                    WHERE w.id = l.wt_sequence_id
+               """
+    )
+
+    op.alter_column("wild_type_sequences", "reference_id", nullable=True)
+    op.drop_table("reference_maps")
+
+
+def downgrade():
+    op.create_table(
+        "reference_maps",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("is_primary", sa.Boolean(), nullable=False),
+        sa.Column("genome_id", sa.Integer(), sa.ForeignKey("reference_genomes.id"), nullable=False),
+        sa.Column("target_id", sa.Integer(), sa.ForeignKey("target_genes.id"), nullable=False),
+        sa.Column("creation_date", sa.Date(), nullable=False),
+        sa.Column("modification_date", sa.Date(), nullable=False),
+        sa.PrimaryKeyConstraint("id"),
+    )
+    op.execute("delete from target_genes where wt_sequence_id is NULL")
+    op.execute(
+        """
+                INSERT INTO reference_maps (is_primary, genome_id, target_id, creation_date, modification_date)
+                    SELECT false, wild_type_sequences.reference_id, target_genes.id, current_date, current_date
+                    FROM target_genes JOIN wild_type_sequences on target_genes.wt_sequence_id = wild_type_sequences.id
+                    WHERE wild_type_sequences.reference_id is not null
+                """
+    )
+    op.execute(
+        "delete from reference_maps where target_id in (select id from target_genes where wt_sequence_id is NULL)"
+    )
+    op.drop_column("wild_type_sequences", "reference_id")

alembic/versions/60103ad1cb5b_add_target_sequence_label.py

@@ -0,0 +1,24 @@
+"""Add Target Sequence Label
+
+Revision ID: 60103ad1cb5b
+Revises: 194cfebabe32
+Create Date: 2023-08-29 16:04:44.620385
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "60103ad1cb5b"
+down_revision = "194cfebabe32"
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    op.add_column("target_sequences", sa.Column("label", sa.String(), nullable=True))
+
+
+def downgrade():
+    op.drop_column("target_sequences", "label")

alembic/versions/90e7860964a2_add_target_accession.py

@@ -0,0 +1,42 @@
+"""Add Target Accession
+
+Revision ID: 90e7860964a2
+Revises: 22e2d92d602e
+Create Date: 2023-08-04 11:40:37.434740
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "90e7860964a2"
+down_revision = "22e2d92d602e"
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.create_table(
+        "target_accessions",
+        sa.Column("id", sa.Integer(), nullable=False),
+        sa.Column("assembly", sa.String(), nullable=False),
+        sa.Column("accession", sa.String(), nullable=False),
+        sa.Column("creation_date", sa.Date(), nullable=False),
+        sa.Column("modification_date", sa.Date(), nullable=False),
+        sa.PrimaryKeyConstraint("id"),
+    )
+    op.add_column(
+        "target_genes", sa.Column("accession_id", sa.Integer, sa.ForeignKey("target_accessions.id"), nullable=True)
+    )
+    op.alter_column("target_genes", "wt_sequence_id", nullable=True)
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column("target_genes", "accession_id")
+    op.drop_table("target_accessions")
+    op.alter_column("target_genes", "wt_sequence_id", nullable=False)
+    # ### end Alembic commands ###

alembic/versions/c6154dd7d9b9_add_gene_name_column_to_target_.py

@@ -0,0 +1,30 @@
+"""Add gene name column to target accession definition
+
+Revision ID: c6154dd7d9b9
+Revises: 60103ad1cb5b
+Create Date: 2023-10-04 17:00:42.960917
+
+"""
+from alembic import op
+import sqlalchemy as sa
+
+
+# revision identifiers, used by Alembic.
+revision = "c6154dd7d9b9"
+down_revision = "60103ad1cb5b"
+branch_labels = None
+depends_on = None
+
+
+def upgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column("target_accessions", sa.Column("gene", sa.String(), nullable=True))
+    op.alter_column("target_accessions", "assembly", nullable=True)
+    # ### end Alembic commands ###
+
+
+def downgrade():
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.drop_column("target_accessions", "gene")
+    op.alter_column("target_accessions", "assembly", nullable=False)
+    # ### end Alembic commands ###

docker-compose-dev.yml

@@ -8,6 +8,7 @@ services:
     command: bash -c "uvicorn mavedb.server_main:app --host 0.0.0.0 --port 8000 --reload"
     depends_on:
       - db
+      - seqrepo
     env_file:
       - settings/.env.dev
     environment:
@@ -17,6 +18,7 @@ services:
       - "8002:8000"
     volumes:
       - .:/code
+      - mavedb-seqrepo-dev:/usr/local/share/seqrepo
 
   db:
     image: postgres:14
@@ -27,10 +29,16 @@ services:
     volumes:
       - mavedb-data-dev:/var/lib/postgresql/data
 
+  seqrepo:
+    image: biocommons/seqrepo:latest
+    volumes:
+      - mavedb-seqrepo-dev:/usr/local/share/seqrepo
+
 #  rabbitmq:
 #    image: rabbitmq:3.8.3
 #    ports:
 #      - "5673:5672"
 
 volumes:
   mavedb-data-dev:
+  mavedb-seqrepo-dev:

pyproject.toml

@@ -23,6 +23,8 @@ dependencies = [
     "IDUtils~=1.2.0",
     "mavehgvs~=0.6.0",
     "eutils~=0.6.0",
+    "hgvs~=1.5.4",
+    "biocommons~=0.0.0",
     "numpy~=1.22.3",
     "pandas~=1.4.1",
     "pydantic~=1.9.1",

requirements.txt

@@ -1,12 +1,14 @@
 alembic~=1.7.6
 authlib~=0.15.5
+biocommons~=0.0.0
 celery~=5.2.3
 cryptography~=37.0.4
 eutils~=0.6.0
 fastapi~=0.71.0
 # fastapi-oidc~=0.0.9
 fqfa~=1.2.3
 IDUtils~=1.2.0
+hgvs~=1.5.4
 jsonschema~=4.17.3
 mavehgvs~=0.6.0
 numpy~=1.22.3

src/mavedb/lib/exceptions.py

@@ -148,3 +148,9 @@ class NonexistentIdentifierError(ValueError):
     """Raised when a user tries to create a publication with a non-existent identifier"""
 
     pass
+
+
+class MixedTargetError(ValueError):
+    """Raised when a user tries to create a score-set against a set of mixed targets"""
+
+    pass

src/mavedb/lib/score_sets.py

@@ -22,8 +22,8 @@
 from mavedb.models.keyword import Keyword
 from mavedb.models.publication_identifier import PublicationIdentifier
 from mavedb.models.reference_genome import ReferenceGenome
-from mavedb.models.reference_map import ReferenceMap
 from mavedb.models.score_set import ScoreSet
+from mavedb.models.target_accession import TargetAccession
 from mavedb.models.target_gene import TargetGene
 from mavedb.models.user import User
 from mavedb.view_models.search import ScoreSetsSearch
@@ -52,8 +52,8 @@ def search_score_sets(db: Session, owner: Optional[User], search: ScoreSetsSearc
                 ScoreSet.title.contains(lower_search_text),
                 ScoreSet.short_description.contains(lower_search_text),
                 ScoreSet.abstract_text.contains(lower_search_text),
-                ScoreSet.target_gene.has(func.lower(TargetGene.name).contains(lower_search_text)),
-                ScoreSet.target_gene.has(func.lower(TargetGene.category).contains(lower_search_text)),
+                ScoreSet.target_genes.any(func.lower(TargetGene.name).contains(lower_search_text)),
+                ScoreSet.target_genes.any(func.lower(TargetGene.category).contains(lower_search_text)),
                 ScoreSet.keyword_objs.any(func.lower(Keyword.text).contains(lower_search_text)),
                 # TODO Add: ORGANISM_NAME UNIPROT, ENSEMBL, REFSEQ, LICENSE, plus TAX_ID if numeric
                 ScoreSet.publication_identifiers.any(
@@ -77,19 +77,17 @@ def search_score_sets(db: Session, owner: Optional[User], search: ScoreSetsSearc
         )
 
     if search.targets:
-        query = query.filter(ScoreSet.target_gene.has(TargetGene.name.in_(search.targets)))
+        query = query.filter(ScoreSet.target_genes.any(TargetGene.name.in_(search.targets)))
 
     if search.target_organism_names:
         query = query.filter(
-            ScoreSet.target_gene.has(
-                TargetGene.reference_maps.any(
-                    ReferenceMap.genome.has(ReferenceGenome.organism_name.in_(search.target_organism_names))
-                )
+            ScoreSet.target_genes.any(
+                TargetGene.reference.any(ReferenceGenome.organism_name.in_(search.target_organism_names))
             )
         )
 
     if search.target_types:
-        query = query.filter(ScoreSet.target_gene.has(TargetGene.category.in_(search.target_types)))
+        query = query.filter(ScoreSet.target_genes.any(TargetGene.category.in_(search.target_types)))
 
     if search.publication_identifiers:
         query = query.filter(
@@ -111,8 +109,15 @@ def search_score_sets(db: Session, owner: Optional[User], search: ScoreSetsSearc
             )
         )
 
+    if search.target_accessions:
+        query = query.filter(
+            ScoreSet.target_genes.any(
+                TargetGene.target_accession.has(TargetAccession.accession.in_(search.target_accessions))
+            )
+        )
+
     score_sets: list[ScoreSet] = (
-        query.join(ScoreSet.experiment).join(ScoreSet.target_gene).order_by(Experiment.title).all()
+        query.join(ScoreSet.experiment).join(ScoreSet.target_genes).order_by(Experiment.title).all()
     )
     if not score_sets:
         score_sets = []