cool-seq-tool 0.4.1__py3-none-any.whl → 0.5.1__py3-none-any.whl

cool_seq_tool/mappers/mane_transcript.py

@@ -21,6 +21,7 @@ import polars as pl
 from pydantic import BaseModel
 
 from cool_seq_tool.handlers.seqrepo_access import SeqRepoAccess
+from cool_seq_tool.mappers.liftover import LiftOver
 from cool_seq_tool.schemas import (
     AnnotationLayer,
     Assembly,
@@ -35,7 +36,7 @@ from cool_seq_tool.sources import (
 )
 from cool_seq_tool.utils import get_inter_residue_pos
 
-logger = logging.getLogger(__name__)
+_logger = logging.getLogger(__name__)
 
 
 class EndAnnotationLayer(str, Enum):
@@ -92,6 +93,7 @@ class ManeTranscript:
         transcript_mappings: TranscriptMappings,
         mane_transcript_mappings: ManeTranscriptMappings,
         uta_db: UtaDatabase,
+        liftover: LiftOver,
     ) -> None:
         """Initialize the ManeTranscript class.
 
@@ -117,11 +119,13 @@ class ManeTranscript:
         :param mane_transcript_mappings: Access to MANE Transcript accession mapping
             data
         :param uta_db: UtaDatabase instance to give access to query UTA database
+        :param liftover: Instance to provide mapping between human genome assemblies
         """
         self.seqrepo_access = seqrepo_access
         self.transcript_mappings = transcript_mappings
         self.mane_transcript_mappings = mane_transcript_mappings
         self.uta_db = uta_db
+        self.liftover = liftover
 
     @staticmethod
     def _get_reading_frame(pos: int) -> int:
@@ -168,10 +172,10 @@ class ManeTranscript:
                 elif ac.startswith("ENSP"):
                     ac = self.transcript_mappings.ensp_to_enst[ac]
                 else:
-                    logger.warning("Unable to find accession: %s", ac)
+                    _logger.warning("Unable to find accession: %s", ac)
                     return None
             except KeyError:
-                logger.warning("%s not found in transcript_mappings", ac)
+                _logger.warning("%s not found in transcript_mappings", ac)
                 return None
 
         pos = self._p_to_c_pos(start_pos, end_pos)
@@ -196,7 +200,7 @@ class ManeTranscript:
                     0
                 ]
             ):
-                logger.warning("Ensembl transcript not found: %s", ac)
+                _logger.warning("Ensembl transcript not found: %s", ac)
                 return None
 
             temp_ac = ac.split(".")[0]
@@ -206,7 +210,7 @@ class ManeTranscript:
         # c. coordinate does not contain cds start, so we need to add it
         cds_start_end = await self.uta_db.get_cds_start_end(temp_ac)
         if not cds_start_end:
-            logger.warning("Accession %s not found in UTA", temp_ac)
+            _logger.warning("Accession %s not found in UTA", temp_ac)
             return None
         coding_start_site = cds_start_end[0]
         pos = pos[0] + coding_start_site, pos[1] + coding_start_site
@@ -215,6 +219,99 @@ class ManeTranscript:
             ac, pos, AnnotationLayer.CDNA, coding_start_site=coding_start_site
         )
 
+    async def _liftover_to_38(self, genomic_tx_data: dict) -> None:
+        """Liftover genomic_tx_data to hg38 assembly.
+
+        :param genomic_tx_data: Dictionary containing gene, nc_accession, alt_pos, and
+            strand. This will be mutated in-place if not GRCh38 assembly.
+        """
+        descr = await self.uta_db.get_chr_assembly(genomic_tx_data["alt_ac"])
+        if descr is None:
+            # already grch38
+            return
+        chromosome, _ = descr
+
+        query = f"""
+            SELECT DISTINCT alt_ac
+            FROM {self.uta_db.schema}.tx_exon_aln_v
+            WHERE tx_ac = '{genomic_tx_data['tx_ac']}';
+            """  # noqa: S608
+        nc_acs = await self.uta_db.execute_query(query)
+        nc_acs = [nc_ac[0] for nc_ac in nc_acs]
+        if nc_acs == [genomic_tx_data["alt_ac"]]:
+            _logger.warning(
+                "UTA does not have GRCh38 assembly for %s",
+                genomic_tx_data["alt_ac"].split(".")[0],
+            )
+            return
+
+        # Get most recent assembly version position
+        # Liftover range
+        self._set_liftover(
+            genomic_tx_data, "alt_pos_range", chromosome, Assembly.GRCH38
+        )
+
+        # Liftover changes range
+        self._set_liftover(
+            genomic_tx_data, "alt_pos_change_range", chromosome, Assembly.GRCH38
+        )
+
+        # Change alt_ac to most recent
+        if genomic_tx_data["alt_ac"].startswith("EN"):
+            order_by_cond = "ORDER BY alt_ac DESC;"
+        else:
+            order_by_cond = """
+            ORDER BY CAST(SUBSTR(alt_ac, position('.' in alt_ac) + 1,
+            LENGTH(alt_ac)) AS INT) DESC;
+            """
+        query = f"""
+            SELECT alt_ac
+            FROM {self.uta_db.schema}.genomic
+            WHERE alt_ac LIKE '{genomic_tx_data['alt_ac'].split('.')[0]}%'
+            {order_by_cond}
+            """  # noqa: S608
+        nc_acs = await self.uta_db.execute_query(query)
+        genomic_tx_data["alt_ac"] = nc_acs[0][0]
+
+    def _set_liftover(
+        self,
+        genomic_tx_data: dict,
+        key: str,
+        chromosome: str,
+        liftover_to_assembly: Assembly,
+    ) -> None:
+        """Update genomic_tx_data to have coordinates for given assembly.
+
+        :param genomic_tx_data: Dictionary containing gene, nc_accession, alt_pos, and
+            strand
+        :param key: Key to access coordinate positions
+        :param chromosome: Chromosome, must be prefixed with ``chr``
+        :param liftover_to_assembly: Assembly to liftover to
+        """
+        liftover_start_i = self.liftover.get_liftover(
+            chromosome, genomic_tx_data[key][0], liftover_to_assembly
+        )
+        if liftover_start_i is None:
+            _logger.warning(
+                "Unable to liftover position %s on %s",
+                genomic_tx_data[key][0],
+                chromosome,
+            )
+            return
+
+        liftover_end_i = self.liftover.get_liftover(
+            chromosome, genomic_tx_data[key][1], liftover_to_assembly
+        )
+        if liftover_end_i is None:
+            _logger.warning(
+                "Unable to liftover position %s on %s",
+                genomic_tx_data[key][1],
+                chromosome,
+            )
+            return
+
+        genomic_tx_data[key] = liftover_start_i[1], liftover_end_i[1]
+
     async def _get_and_validate_genomic_tx_data(
         self,
         tx_ac: str,
@@ -237,7 +334,7 @@ class ManeTranscript:
             tx_ac, pos, annotation_layer, alt_ac=alt_ac
         )
         if not genomic_tx_data:
-            logger.warning(
+            _logger.warning(
                 "Unable to find genomic_tx_data for %s at position %s on annotation layer %s",
                 alt_ac,
                 pos,
@@ -250,12 +347,12 @@ class ManeTranscript:
             # Only want to liftover if alt_ac not provided. If alt_ac is provided,
             # it means user wants to stick with the queried assembly
             og_alt_exon_id = genomic_tx_data["alt_exon_id"]
-            await self.uta_db.liftover_to_38(genomic_tx_data)
+            await self._liftover_to_38(genomic_tx_data)
             liftover_alt_exon_id = genomic_tx_data["alt_exon_id"]
 
             # Validation check: Exon structure
             if og_alt_exon_id != liftover_alt_exon_id:
-                logger.warning(
+                _logger.warning(
                     "Original alt_exon_id %s does not match liftover alt_exon_id %s",
                     og_alt_exon_id,
                     liftover_alt_exon_id,
@@ -293,7 +390,7 @@ class ManeTranscript:
         gt_cds_end = c_pos_change[1] > cds_end and c_pos_change[1] > cds_end
 
         if lt_cds_start or gt_cds_end:
-            logger.info(
+            _logger.info(
                 "%s with position %s is not within CDS start/end",
                 refseq_c_ac,
                 c_pos_change,
@@ -381,7 +478,7 @@ class ManeTranscript:
             )
 
             if not result:
-                logger.warning(
+                _logger.warning(
                     "Unable to find transcript, %s, position change", refseq_c_ac
                 )
                 return None
@@ -438,7 +535,7 @@ class ManeTranscript:
                 new_rf = self._get_reading_frame(transcript_data.pos[pos_index])
 
                 if og_rf != new_rf:
-                    logger.warning(
+                    _logger.warning(
                         "%s original reading frame (%s) does not match new %s, %s reading frame (%s)",
                         ac,
                         og_rf,
@@ -449,7 +546,7 @@ class ManeTranscript:
                     return False
             else:
                 if pos_index == 0:
-                    logger.warning("%s must having start position", ac)
+                    _logger.warning("%s must having start position", ac)
                     return False
         return True
 
@@ -503,10 +600,10 @@ class ManeTranscript:
                 residue_mode=residue_mode,
             )
             if not mane_ref:
-                logger.info("Unable to validate reference for MANE Transcript")
+                _logger.info("Unable to validate reference for MANE Transcript")
 
             if expected_ref != mane_ref:
-                logger.info(
+                _logger.info(
                     "Expected ref, %s, but got %s on MANE accession, %s",
                     expected_ref,
                     mane_ref,
@@ -514,7 +611,7 @@ class ManeTranscript:
                 )
 
         if expected_ref != ref:
-            logger.warning(
+            _logger.warning(
                 "Expected ref, %s, but got %s on accession, %s", expected_ref, ref, ac
             )
             return False
@@ -692,7 +789,7 @@ class ManeTranscript:
             )
 
         if df.is_empty():
-            logger.warning("Unable to get transcripts from gene %s", gene)
+            _logger.warning("Unable to get transcripts from gene %s", gene)
             return lcr_result
 
         prioritized_tx_acs = self._get_prioritized_transcripts_from_gene(df)
@@ -814,7 +911,7 @@ class ManeTranscript:
                 pos = lcr_result.pos
 
                 if not self._validate_index(ac, pos, coding_start_site):
-                    logger.warning(
+                    _logger.warning(
                         "%s are not valid positions on %s with coding start site %s",
                         pos,
                         ac,
@@ -841,7 +938,7 @@ class ManeTranscript:
                 pos = lcr_result_dict[k]["pos"]
                 if not self._validate_index(ac, pos, cds):
                     valid = False
-                    logger.warning(
+                    _logger.warning(
                         "%s are not valid positions on %s with coding start site %s",
                         pos,
                         ac,
@@ -999,7 +1096,7 @@ class ManeTranscript:
             return await self.g_to_mane_c(
                 ac, start_pos, end_pos, gene=gene, residue_mode=residue_mode
             )
-        logger.warning("Annotation layer not supported: %s", start_annotation_layer)
+        _logger.warning("Annotation layer not supported: %s", start_annotation_layer)
         return None
 
     async def g_to_grch38(self, ac: str, start_pos: int, end_pos: int) -> dict | None:
@@ -1024,11 +1121,11 @@ class ManeTranscript:
         is_same_pos = start_pos == end_pos
 
         # Coordinate liftover
-        if assembly < "GRCh37":
-            logger.warning("Liftover only supported for GRCh37")
+        if assembly < Assembly.GRCH37:
+            _logger.warning("Liftover only supported for GRCh37")
             return None
 
-        liftover_start_i = self.uta_db.get_liftover(
+        liftover_start_i = self.liftover.get_liftover(
             chromosome, start_pos, Assembly.GRCH38
         )
         if liftover_start_i is None:
@@ -1036,7 +1133,7 @@ class ManeTranscript:
         start_pos = liftover_start_i[1]
 
         if not is_same_pos:
-            liftover_end_i = self.uta_db.get_liftover(
+            liftover_end_i = self.liftover.get_liftover(
                 chromosome, end_pos, Assembly.GRCH38
             )
             if liftover_end_i is None:
@@ -1130,7 +1227,7 @@ class ManeTranscript:
             )
 
         if not await self.uta_db.validate_genomic_ac(ac):
-            logger.warning("Genomic accession does not exist: %s", ac)
+            _logger.warning("Genomic accession does not exist: %s", ac)
             return None
 
         mane_data = self.mane_transcript_mappings.get_gene_mane_data(gene)
@@ -1156,7 +1253,7 @@ class ManeTranscript:
                 )
                 if not mane_tx_genomic_data:
                     continue
-                logger.info("Not using most recent assembly")
+                _logger.info("Not using most recent assembly")
 
             coding_start_site = mane_tx_genomic_data["coding_start_site"]
             coding_end_site = mane_tx_genomic_data["coding_end_site"]
@@ -1167,7 +1264,7 @@ class ManeTranscript:
             if not self._validate_index(
                 mane_c_ac, mane_c_pos_change, coding_start_site
             ):
-                logger.warning(
+                _logger.warning(
                     "%s are not valid positions on %s with coding start site %s",
                     mane_c_pos_change,
                     mane_c_ac,
@@ -1257,7 +1354,7 @@ class ManeTranscript:
             if not self._validate_index(
                 mane_c_ac, mane_c_pos_change, coding_start_site
             ):
-                logger.warning(
+                _logger.warning(
                     "%s are not valid positions on %s with coding start site %s",
                     mane_c_pos_change,
                     mane_c_ac,

cool_seq_tool/resources/status.py

@@ -9,8 +9,9 @@ from asyncpg import InvalidCatalogNameError, UndefinedTableError
 from biocommons.seqrepo import SeqRepo
 
 from cool_seq_tool.handlers.seqrepo_access import SEQREPO_ROOT_DIR, SeqRepoAccess
+from cool_seq_tool.mappers.liftover import LiftOver
 from cool_seq_tool.resources.data_files import DataFile, get_data_file
-from cool_seq_tool.sources.uta_database import UTA_DB_URL, UtaDatabase, get_liftover
+from cool_seq_tool.sources.uta_database import UTA_DB_URL, UtaDatabase
 
 _logger = logging.getLogger(__name__)
 
@@ -42,9 +43,7 @@ async def check_status(
     Arguments are intended to mirror arguments to :py:meth:`cool_seq_tool.app.CoolSeqTool.__init__`.
 
     Additional arguments are available for testing paths to specific chainfiles (same
-    signature as :py:meth:`cool_seq_tool.sources.uta_database.UtaDatabase.__init__`).
-    Note that chainfile failures also entail UTA initialization failure; this status is
-    reported separately to enable more precise debugging.
+    signature as :py:meth:`cool_seq_tool.mappers.liftover.LiftOver.__init__`).
 
     >>> from cool_seq_tool.resources.status import check_status
     >>> await check_status()
@@ -104,7 +103,10 @@ async def check_status(
             status[name_lower] = True
 
     try:
-        get_liftover(chain_file_37_to_38, chain_file_38_to_37)
+        LiftOver(
+            chain_file_37_to_38=chain_file_37_to_38,
+            chain_file_38_to_37=chain_file_38_to_37,
+        )
     except (FileNotFoundError, ChainfileError) as e:
         _logger.error("agct converter setup failed: %s", e)
     except Exception as e:

cool_seq_tool/schemas.py

@@ -1,7 +1,6 @@
 """Defines attribute constants, useful object structures, and API response schemas."""
 
 import datetime
-import re
 from enum import Enum, IntEnum
 from typing import Literal
 
@@ -10,11 +9,10 @@ from pydantic import (
     ConfigDict,
     StrictInt,
     StrictStr,
-    field_validator,
     model_validator,
 )
 
-from cool_seq_tool.version import __version__
+from cool_seq_tool import __version__
 
 _now = str(datetime.datetime.now(tz=datetime.timezone.utc))
 
@@ -35,11 +33,16 @@ class Strand(IntEnum):
 
 
 class Assembly(str, Enum):
-    """Create Enum for supported genomic assemblies"""
+    """Define supported genomic assemblies. Must be defined in ascending order"""
 
     GRCH37 = "GRCh37"
     GRCH38 = "GRCh38"
 
+    @classmethod
+    def values(cls) -> list[str]:
+        """Return list of values in enum (ascending assembly order)"""
+        return [item.value for item in cls]
+
 
 class TranscriptPriority(str, Enum):
     """Create Enum for Transcript Priority labels"""
@@ -276,17 +279,6 @@ class ServiceMeta(BaseModelForbidExtra):
         "https://github.com/GenomicMedLab/cool-seq-tool"
     )
 
-    @field_validator("version")
-    def validate_version(cls, v):
-        """Check version matches semantic versioning regex pattern.
-        https://semver.org/#is-there-a-suggested-regular-expression-regex-to-check-a-semver-string
-        """
-        version_regex = r"^(0|[1-9]\d*)\.(0|[1-9]\d*)\.(0|[1-9]\d*)(?:-((?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*)(?:\.(?:0|[1-9]\d*|\d*[a-zA-Z-][0-9a-zA-Z-]*))*))?(?:\+([0-9a-zA-Z-]+(?:\.[0-9a-zA-Z-]+)*))?$"
-        if not re.match(version_regex, v):
-            msg = f"Invalid version {v}"
-            raise ValueError(msg)
-        return v
-
     model_config = ConfigDict(
         json_schema_extra={
             "example": {
@@ -384,7 +376,7 @@ class MappedManeData(BaseModel):
                 "strand": Strand.NEGATIVE,
                 "status": TranscriptPriority.MANE_PLUS_CLINICAL,
                 "alt_ac": "NC_000007.13",
-                "assembly": "GRCh37",
+                "assembly": Assembly.GRCH37,
             }
         }
     )
@@ -407,7 +399,7 @@ class MappedManeDataService(BaseModelForbidExtra):
                     "strand": Strand.NEGATIVE,
                     "status": TranscriptPriority.MANE_PLUS_CLINICAL,
                     "alt_ac": "NC_000007.13",
-                    "assembly": "GRCh37",
+                    "assembly": Assembly.GRCH37,
                 },
                 "warnings": [],
                 "service_meta": {

cool_seq_tool/sources/mane_transcript_mappings.py

@@ -9,7 +9,7 @@ import polars as pl
 
 from cool_seq_tool.resources.data_files import DataFile, get_data_file
 
-logger = logging.getLogger(__name__)
+_logger = logging.getLogger(__name__)
 
 
 class ManeTranscriptMappings:
@@ -63,7 +63,7 @@ class ManeTranscriptMappings:
         data = self.df.filter(pl.col("symbol") == gene_symbol.upper())
 
         if len(data) == 0:
-            logger.warning(
+            _logger.warning(
                 "Unable to get MANE Transcript data for gene: %s", gene_symbol
             )
             return []