cool-seq-tool 0.14.1__py3-none-any.whl → 0.14.2__py3-none-any.whl

cool_seq_tool/mappers/__init__.py

@@ -4,6 +4,13 @@ from .alignment import AlignmentMapper  # noqa: I001
 from .liftover import LiftOver
 from .mane_transcript import ManeTranscript
 from .exon_genomic_coords import ExonGenomicCoordsMapper
+from .feature_overlap import FeatureOverlap
 
 
-__all__ = ["AlignmentMapper", "ExonGenomicCoordsMapper", "LiftOver", "ManeTranscript"]
+__all__ = [
+    "AlignmentMapper",
+    "ExonGenomicCoordsMapper",
+    "FeatureOverlap",
+    "LiftOver",
+    "ManeTranscript",
+]

cool_seq_tool/mappers/feature_overlap.py

@@ -0,0 +1,251 @@
+"""Module for getting feature (gene/exon) overlap"""
+
+import re
+from pathlib import Path
+
+import polars as pl
+from ga4gh.core import ga4gh_identify
+from ga4gh.vrs.models import SequenceLocation, SequenceReference
+
+from cool_seq_tool.handlers import SeqRepoAccess
+from cool_seq_tool.resources.data_files import DataFile, get_data_file
+from cool_seq_tool.schemas import Assembly, CdsOverlap, CoordinateType
+
+# Pattern for chromosome
+CHR_PATTERN = r"X|Y|([1-9]|1[0-9]|2[0-2])"
+
+
+class FeatureOverlapError(Exception):
+    """Custom exception for the Feature Overlap class"""
+
+
+class FeatureOverlap:
+    """The class for getting feature overlap"""
+
+    def __init__(
+        self,
+        seqrepo_access: SeqRepoAccess,
+        mane_refseq_genomic_path: Path | None = None,
+        from_local: bool = False,
+    ) -> None:
+        """Initialize the FeatureOverlap class. Will load RefSeq data and store as df.
+
+        :param seqrepo_access: Client for accessing SeqRepo data
+        :param mane_refseq_genomic_path: Path to MANE RefSeq Genomic GFF data
+        :param from_local: if ``True``, don't check for or acquire latest version --
+            just provide most recent locally available file, if possible, and raise
+            error otherwise
+        """
+        if not mane_refseq_genomic_path:
+            mane_refseq_genomic_path = get_data_file(
+                DataFile.MANE_REFSEQ_GENOMIC, from_local
+            )
+        self.seqrepo_access = seqrepo_access
+        self.mane_refseq_genomic_path = mane_refseq_genomic_path
+        self.df = self._load_mane_refseq_gff_data()
+
+    def _load_mane_refseq_gff_data(self) -> pl.DataFrame:
+        """Load MANE RefSeq GFF data file into DataFrame.
+
+        :return: DataFrame containing MANE RefSeq Genomic GFF data for CDS. Columns
+            include `type`, `chromosome` (chromosome without 'chr' prefix), `cds_start`,
+            `cds_stop`, `info_name` (name of record), and `gene`. `cds_start` and
+            `cds_stop` use inter-residue coordinates.
+        """
+        df = pl.read_csv(
+            self.mane_refseq_genomic_path,
+            separator="\t",
+            has_header=False,
+            skip_rows=9,
+            columns=[0, 2, 3, 4, 8],
+        )
+        df.columns = ["chromosome", "type", "cds_start", "cds_stop", "info"]
+
+        # Restrict to only feature of interest: CDS (which has gene info)
+        df = df.filter(pl.col("type") == "CDS")
+
+        # Get name from the info field
+        # Get gene from the info field
+        # Get chromosome names without prefix and without suffix for alternate transcripts
+        # Convert start and stop to ints
+        # Convert to inter-residue coordinates
+        # Only return certain columns
+        return df.with_columns(
+            (pl.col("info").str.extract(r"Name=([^;]+)", 1).alias("info_name")),
+            (pl.col("info").str.extract(r"gene=([^;]+)", 1).alias("gene")),
+            (pl.col("chromosome").str.extract(r"^chr?([^_]+)", 1).alias("chromosome")),
+            (pl.col("cds_start").cast(pl.Int64) - 1).alias("cds_start"),
+            (pl.col("cds_stop").cast(pl.Int64).alias("cds_stop")),
+        ).select(
+            [
+                pl.col("type"),
+                pl.col("chromosome"),
+                pl.col("cds_start"),
+                pl.col("cds_stop"),
+                pl.col("info_name"),
+                pl.col("gene"),
+            ]
+        )
+
+    def _get_chr_from_alt_ac(self, identifier: str) -> str:
+        """Get chromosome given genomic identifier
+
+        :param identifier: Genomic identifier on GRCh38 assembly
+        :raises FeatureOverlapError: If unable to find associated GRCh38 chromosome
+        :return: Chromosome. 1..22, X, Y. No 'chr' prefix.
+        """
+        aliases, error_msg = self.seqrepo_access.translate_identifier(
+            identifier, Assembly.GRCH38.value
+        )
+
+        if error_msg:
+            raise FeatureOverlapError(str(error_msg))
+
+        if not aliases:
+            error_msg = (
+                f"Unable to find {Assembly.GRCH38.value} aliases for: {identifier}"
+            )
+            raise FeatureOverlapError(error_msg)
+
+        assembly_chr_pattern = (
+            rf"^{Assembly.GRCH38.value}:(?P<chromosome>{CHR_PATTERN})$"
+        )
+        for a in aliases:
+            chr_match = re.match(assembly_chr_pattern, a)
+            if chr_match:
+                break
+
+        if not chr_match:
+            error_msg = (
+                f"Unable to find {Assembly.GRCH38.value} chromosome for: {identifier}"
+            )
+            raise FeatureOverlapError(error_msg)
+
+        chr_groupdict = chr_match.groupdict()
+        return chr_groupdict["chromosome"]
+
+    def get_grch38_mane_gene_cds_overlap(
+        self,
+        start: int,
+        end: int,
+        chromosome: str | None = None,
+        identifier: str | None = None,
+        coordinate_type: CoordinateType = CoordinateType.RESIDUE,
+    ) -> dict[str, list[CdsOverlap]] | None:
+        """Given GRCh38 genomic data, find the overlapping MANE features (gene and cds).
+        The genomic data is specified as a sequence location by `chromosome`, `start`,
+        `end`. All CDS regions with which the input sequence location has nonzero base
+        pair overlap will be returned.
+
+        :param start: GRCh38 start position
+        :param end: GRCh38 end position
+        :param chromosome: Chromosome. 1..22, X, or Y. If not provided, must provide
+            `identifier`. If both `chromosome` and `identifier` are provided,
+            `chromosome` will be used.
+        :param identifier: Genomic identifier on GRCh38 assembly. If not provided, must
+            provide `chromosome`. If both `chromosome` and `identifier` are provided,
+            `chromosome` will be used.
+        :param coordinate_type: Coordinate type for ``start`` and ``end``
+        :raise FeatureOverlapError: If missing required fields or unable to find
+            associated ga4gh identifier
+        :return: MANE feature (gene/cds) overlap data represented as a dict. The
+            dictionary will be keyed by genes which overlap the input sequence location.
+            Each gene contains a list of the overlapping CDS regions with the beginning
+            and end of the input sequence location's overlap with each
+        """
+        ga4gh_seq_id = None
+        if chromosome:
+            if not re.match(f"^{CHR_PATTERN}$", chromosome):
+                error_msg = "`chromosome` must be 1, ..., 22, X, or Y"
+                raise FeatureOverlapError(error_msg)
+        else:
+            if identifier:
+                chromosome = self._get_chr_from_alt_ac(identifier)
+                if identifier.startswith("ga4gh:SQ."):
+                    ga4gh_seq_id = identifier
+            else:
+                error_msg = "Must provide either `chromosome` or `identifier`"
+                raise FeatureOverlapError(error_msg)
+
+        # Convert residue to inter-residue
+        if coordinate_type == CoordinateType.RESIDUE:
+            start -= 1
+
+        # Get feature dataframe (df uses inter-residue)
+        feature_df = self.df.filter(
+            (pl.col("chromosome") == chromosome)
+            & (pl.col("cds_start") <= end)
+            & (pl.col("cds_stop") >= start)
+        )
+
+        if feature_df.is_empty():
+            return None
+
+        # Add overlap columns
+        feature_df = feature_df.with_columns(
+            [
+                pl.when(pl.col("cds_start") < start)
+                .then(start)
+                .otherwise(pl.col("cds_start"))
+                .alias("overlap_start"),
+                pl.when(pl.col("cds_stop") > end)
+                .then(end)
+                .otherwise(pl.col("cds_stop"))
+                .alias("overlap_stop"),
+            ]
+        )
+
+        # Get ga4gh identifier for chromosome
+        if not ga4gh_seq_id:
+            grch38_chr = f"{Assembly.GRCH38.value}:{chromosome}"
+            ga4gh_aliases, error_msg = self.seqrepo_access.translate_identifier(
+                grch38_chr, "ga4gh"
+            )
+
+            # Errors should never happen but catching just in case
+            if error_msg:
+                raise FeatureOverlapError(str(error_msg))
+
+            if not ga4gh_aliases:
+                error_msg = f"Unable to find ga4gh identifier for: {grch38_chr}"
+                raise FeatureOverlapError(error_msg)
+
+            ga4gh_seq_id = ga4gh_aliases[0]
+
+        def _get_seq_loc(start_pos: int, stop_pos: int, refget_ac: str) -> dict:
+            """Get VRS Sequence Location represented as a dict
+
+            :param start_pos: Start position
+            :param stop_pos: Stop position
+            :param refget_ac: Refget Accession (SQ.)
+            :return: VRS Sequence Location represented as dictionary with the ga4gh ID
+                included
+            """
+            _sl = SequenceLocation(
+                sequenceReference=SequenceReference(
+                    refgetAccession=refget_ac,
+                ),
+                start=start_pos,
+                end=stop_pos,
+            )
+            ga4gh_identify(_sl)
+            return _sl.model_dump(exclude_none=True)
+
+        resp = {}
+        refget_ac = ga4gh_seq_id.split("ga4gh:")[-1]
+        for gene, group in feature_df.group_by("gene"):
+            gene = gene[0]
+            _gene_overlap_data = [
+                CdsOverlap(
+                    cds=_get_seq_loc(
+                        cds_row["cds_start"], cds_row["cds_stop"], refget_ac
+                    ),
+                    overlap=_get_seq_loc(
+                        cds_row["overlap_start"], cds_row["overlap_stop"], refget_ac
+                    ),
+                ).model_dump(by_alias=True, exclude_none=True)
+                for cds_row in group.iter_rows(named=True)
+            ]
+            resp[gene] = _gene_overlap_data
+
+        return resp

cool_seq_tool/resources/data_files.py

@@ -6,7 +6,11 @@ from importlib import resources
 from os import environ
 from pathlib import Path
 
-from wags_tails import NcbiLrgRefSeqGeneData, NcbiManeSummaryData
+from wags_tails import (
+    NcbiLrgRefSeqGeneData,
+    NcbiManeRefSeqGenomicData,
+    NcbiManeSummaryData,
+)
 
 _logger = logging.getLogger(__name__)
 
@@ -16,6 +20,7 @@ class DataFile(str, Enum):
 
     TRANSCRIPT_MAPPINGS = "transcript_mappings"
     MANE_SUMMARY = "mane_summary"
+    MANE_REFSEQ_GENOMIC = "mane_refseq_genomic"
     LRG_REFSEQGENE = "lrg_refseqgene"
 
     def lower(self) -> str:
@@ -37,6 +42,12 @@ _resource_acquisition_params = {
             from_local=from_local
         )[0],
     ),
+    DataFile.MANE_REFSEQ_GENOMIC: (
+        "MANE_REFSEQ_GENOMIC_PATH",
+        lambda from_local: NcbiManeRefSeqGenomicData(silent=True).get_latest(
+            from_local=from_local
+        )[0],
+    ),
     DataFile.LRG_REFSEQGENE: (
         "LRG_REFSEQGENE_PATH",
         lambda from_local: NcbiLrgRefSeqGeneData(silent=True).get_latest(
@@ -53,6 +64,7 @@ def get_data_file(resource: DataFile, from_local: bool = False) -> Path:
 
     * ``Resource.TRANSCRIPT_MAPPINGS`` -> ``TRANSCRIPT_MAPPINGS_PATH``
     * ``Resource.MANE_SUMMARY`` -> ``MANE_SUMMARY_PATH``
+    * ``Resource.MANE_REFSEQ_GENOMIC`` -> ``MANE_REFSEQ_GENOMIC_PATH``
     * ``Resource.LRG_REFSEQGENE`` -> ``LRG_REFSEQGENE_PATH``
 
     Otherwise, this function falls back on default expected locations:

cool_seq_tool/schemas.py

@@ -4,6 +4,7 @@ import datetime
 from enum import Enum, IntEnum
 from typing import Literal
 
+from ga4gh.vrs.models import SequenceLocation
 from pydantic import (
     BaseModel,
     ConfigDict,
@@ -167,3 +168,37 @@ class ServiceMeta(BaseModelForbidExtra):
             }
         }
     )
+
+
+class CdsOverlap(BaseModelForbidExtra):
+    """Create model for representing CDS start/stop and Overlap start/stop"""
+
+    cds: SequenceLocation
+    overlap: SequenceLocation
+
+    model_config = ConfigDict(
+        json_schema_extra={
+            "example": {
+                "cds": {
+                    "id": "ga4gh:SL.fYRYzNIAoe6UQF9MT1XaYsFscoU68ZJv",
+                    "type": "SequenceLocation",
+                    "sequenceReference": {
+                        "refgetAccession": "SQ.F-LrLMe1SRpfUZHkQmvkVKFEGaoDeHul",
+                        "type": "SequenceReference",
+                    },
+                    "start": 140726493,
+                    "end": 140726516,
+                },
+                "overlap": {
+                    "id": "ga4gh:SL.fYRYzNIAoe6UQF9MT1XaYsFscoU68ZJv",
+                    "type": "SequenceLocation",
+                    "sequenceReference": {
+                        "refgetAccession": "SQ.F-LrLMe1SRpfUZHkQmvkVKFEGaoDeHul",
+                        "type": "SequenceReference",
+                    },
+                    "start": 140726493,
+                    "end": 140726516,
+                },
+            }
+        }
+    )

{cool_seq_tool-0.14.1.dist-info → cool_seq_tool-0.14.2.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cool_seq_tool
-Version: 0.14.1
+Version: 0.14.2
 Summary: Common Operation on Lots of Sequences Tool
 Author: Kori Kuzma, James Stevenson, Katie Stahl, Alex Wagner
 License: MIT License
@@ -45,15 +45,13 @@ Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: asyncpg
-Requires-Dist: aiofiles
 Requires-Dist: boto3
 Requires-Dist: agct>=0.1.0-dev1
 Requires-Dist: polars~=1.0
-Requires-Dist: hgvs
 Requires-Dist: biocommons.seqrepo
 Requires-Dist: pydantic<3.0,>=2.0
 Requires-Dist: ga4gh.vrs<3.0,>=2.1.3
-Requires-Dist: wags-tails~=0.3.2
+Requires-Dist: wags-tails~=0.4.0
 Requires-Dist: bioutils
 Provides-Extra: dev
 Requires-Dist: pre-commit>=4.2.0; extra == "dev"
@@ -64,7 +62,7 @@ Requires-Dist: ruff==0.12.1; extra == "dev"
 Provides-Extra: tests
 Requires-Dist: pytest; extra == "tests"
 Requires-Dist: pytest-cov; extra == "tests"
-Requires-Dist: pytest-asyncio==0.18.3; extra == "tests"
+Requires-Dist: pytest-asyncio; extra == "tests"
 Requires-Dist: mock; extra == "tests"
 Provides-Extra: docs
 Requires-Dist: sphinx==6.1.3; extra == "docs"

{cool_seq_tool-0.14.1.dist-info → cool_seq_tool-0.14.2.dist-info}/RECORD

@@ -1,24 +1,25 @@
 cool_seq_tool/__init__.py,sha256=pJyVj7Z275BBAwpeFMm-WEn_tp-y1_ihRl1sLc4FFZY,400
 cool_seq_tool/app.py,sha256=vyqlQRffC8sWZXMm-f_f-8WuTTWo3oRNfPUa_qdPV2M,4944
-cool_seq_tool/schemas.py,sha256=D0DsYAR1ZX7RONuc7X4hsPMKcZct7_2LlnE1KKVNre0,4139
+cool_seq_tool/schemas.py,sha256=ueEq7cVKyGWVDvsV9JAzg7Cp2Nni3F9musCkWEtW26g,5365
 cool_seq_tool/utils.py,sha256=kesu7UnOplDzvNBg_G-_m1xMM22979nmsi4yWtweetU,2959
 cool_seq_tool/handlers/__init__.py,sha256=KalQ46vX1MO4SJz2SlspKoIRy1n3c3Vp1t4Y2pIfqow,78
 cool_seq_tool/handlers/seqrepo_access.py,sha256=lRzPc8V0eZJTlefbHuVKeZTEC8-KcyPzpqX7vx3amu8,9118
-cool_seq_tool/mappers/__init__.py,sha256=4_YNwNyw_QrlhRNu1nly8Dezv81XjCIiNa7crVXEh38,305
+cool_seq_tool/mappers/__init__.py,sha256=tavpwkNogg_nF1J_kb6Q9jk7ezqdRz063v7BMZ4koLM,390
 cool_seq_tool/mappers/alignment.py,sha256=nV6PS3mhkQ2MD1GcpNBujBOqd3AKxYSYA9BCusFOa1o,9636
 cool_seq_tool/mappers/exon_genomic_coords.py,sha256=t36NhWo2Rl84dgZY6qO7XFmGpfisjAqC-1ZOTRZxWvg,44757
+cool_seq_tool/mappers/feature_overlap.py,sha256=_MMEuZh48dHDPKliQV2C14bwziScE46uWl9utnAHOZo,9845
 cool_seq_tool/mappers/liftover.py,sha256=lltx9zxfkrb5PHtJlKp3a39JCwPP4e0Zft-mQc1jXL8,3367
 cool_seq_tool/mappers/mane_transcript.py,sha256=2cAYi0Y_gGdPI40weH9Ud1uVBCTKuwMT0M7KFUyjzU0,54365
 cool_seq_tool/resources/__init__.py,sha256=VwUC8YaucTS6SmRirToulZTF6CuvuLQRSxFfSfAovCc,77
-cool_seq_tool/resources/data_files.py,sha256=3lhu28tzlSoTs4vHZNu-hhoAWRrPGuZj_oIjqk2sYQM,3837
+cool_seq_tool/resources/data_files.py,sha256=6d1M5WjeFHdTQpzxqjQ78auQRZvIBVqH8QNCrmRRDXw,4205
 cool_seq_tool/resources/status.py,sha256=9LYSO2mOzVmoSQwllzq1mGChjtDA6j3I0S372N89clA,5683
 cool_seq_tool/resources/transcript_mapping.tsv,sha256=AO3luYQAbFiCoRgiiPXotakb5pAwx1jDCeXpvGdIuac,24138769
 cool_seq_tool/sources/__init__.py,sha256=51QiymeptF7AeVGgV-tW_9f4pIUr0xtYbyzpvHOCneM,304
 cool_seq_tool/sources/mane_transcript_mappings.py,sha256=C5puIA1xuEzBaSvs8VtSxVb2OIDGUg5no8v6Ma2QSdw,6597
 cool_seq_tool/sources/transcript_mappings.py,sha256=903RKTMBO2rbKh6iTQ1BEWnY4C7saBFMPw2_4ATuudg,10054
 cool_seq_tool/sources/uta_database.py,sha256=zzRzmYuybqzEg7zeuQjhK46SPK5GfbiWWNRGNJju8AI,36197
-cool_seq_tool-0.14.1.dist-info/licenses/LICENSE,sha256=IpqC9A-tZW7XXXvCS8c4AVINqkmpxiVA-34Qe3CZSjo,1072
-cool_seq_tool-0.14.1.dist-info/METADATA,sha256=TZHF8lDyidXRdGWUoBWfYAToUdQ7H6yO-gkpUhBdOe4,6587
-cool_seq_tool-0.14.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-cool_seq_tool-0.14.1.dist-info/top_level.txt,sha256=cGuxdN6p3y16jQf6hCwWhE4OptwUeZPm_PNJlPb3b0k,14
-cool_seq_tool-0.14.1.dist-info/RECORD,,
+cool_seq_tool-0.14.2.dist-info/licenses/LICENSE,sha256=IpqC9A-tZW7XXXvCS8c4AVINqkmpxiVA-34Qe3CZSjo,1072
+cool_seq_tool-0.14.2.dist-info/METADATA,sha256=qtjqeUzTDsOnTlil7dVW7Uzg50mQ_-Lu8sKwR95ExGs,6535
+cool_seq_tool-0.14.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+cool_seq_tool-0.14.2.dist-info/top_level.txt,sha256=cGuxdN6p3y16jQf6hCwWhE4OptwUeZPm_PNJlPb3b0k,14
+cool_seq_tool-0.14.2.dist-info/RECORD,,