feat!: remove beacon translator

close #622

* Translator functionality was deprecated in #578 . This commit removes Beacon translator completely

Author: korikuzma

CLAUDE.md

@@ -37,7 +37,7 @@ VRS-Python is the reference implementation of the GA4GH Variation Representation
 - **`src/ga4gh/vrs/enderef.py`** - Converting between inlined and referenced VRS object forms
 
 ### Key Modules
-- **`src/ga4gh/vrs/extras/translator.py`** - Translates between VRS and external formats (HGVS, SPDI, gnomAD, Beacon)
+- **`src/ga4gh/vrs/extras/translator.py`** - Translates between VRS and external formats (HGVS, SPDI, gnomAD)
 - **`src/ga4gh/vrs/extras/annotator/`** - VCF annotation tools with VRS identifiers
 - **`src/ga4gh/vrs/utils/hgvs_tools.py`** - HGVS parsing and validation utilities
 - **`src/ga4gh/core/`** - Core GA4GH models and identifier generation

notebooks/getting_started/4_Exploring_the_AlleleTranslator.ipynb

@@ -11,7 +11,7 @@
    },
    "source": [
     "# 4. Exploring the AlleleTranslator\n",
-    "There are four variant nomenclatures available in the vrs-python *AlleleTranslator*: SPDI, gnomad/VCF, Beacon and HGVS. In this notebook we will perform a simple Allele translation for each. We will use each of the four nomenclatures for a single variant in translating variants to VRS. his variant can be viewed in \n",
+    "There are four variant nomenclatures available in the vrs-python *AlleleTranslator*: SPDI, gnomad/VCF, and HGVS. In this notebook we will perform a simple Allele translation for each. We will use each of the four nomenclatures for a single variant in translating variants to VRS. his variant can be viewed in \n",
     "[ClinVar](https://www.ncbi.nlm.nih.gov/clinvar/variation/652570) and in [gnomAD](https://gnomad.broadinstitute.org/variant/5-80656489-C-T)."
    ]
   },
@@ -386,61 +386,6 @@
     "allele_translator.translate_to(allele, \"spdi\")"
    ]
   },
-  {
-   "cell_type": "markdown",
-   "id": "4e0911a7694a060",
-   "metadata": {
-    "collapsed": false,
-    "jupyter": {
-     "outputs_hidden": false
-    }
-   },
-   "source": [
-    "#### From Beacon (VCF-like)\n",
-    "For variants represented in the Beacon nomenclature, the *AlleleTranslator* currently only supports *translate_from* to convert to VRS. *translate_to* is not yet supported."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 10,
-   "id": "57f54e6c3854a48f",
-   "metadata": {
-    "ExecuteTime": {
-     "end_time": "2024-04-19T17:13:51.743020Z",
-     "start_time": "2024-04-19T17:13:51.663593Z"
-    },
-    "collapsed": false,
-    "jupyter": {
-     "outputs_hidden": false
-    }
-   },
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "{'id': 'ga4gh:VA.ebezGL6HoAhtGJyVnB_mE5BH18ntKev4',\n",
-       " 'type': 'Allele',\n",
-       " 'digest': 'ebezGL6HoAhtGJyVnB_mE5BH18ntKev4',\n",
-       " 'location': {'id': 'ga4gh:SL.JiLRuuyS5wefF_6-Vw7m3Yoqqb2YFkss',\n",
-       "  'type': 'SequenceLocation',\n",
-       "  'digest': 'JiLRuuyS5wefF_6-Vw7m3Yoqqb2YFkss',\n",
-       "  'sequenceReference': {'type': 'SequenceReference',\n",
-       "   'refgetAccession': 'SQ.aUiQCzCPZ2d0csHbMSbh2NzInhonSXwI'},\n",
-       "  'start': 80656488,\n",
-       "  'end': 80656489},\n",
-       " 'state': {'type': 'LiteralSequenceExpression', 'sequence': 'T'}}"
-      ]
-     },
-     "execution_count": 10,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "allele = allele_translator.translate_from(\"5 : 80656489 C > T\", \"beacon\")\n",
-    "allele.model_dump(exclude_none=True)"
-   ]
-  },
   {
    "cell_type": "markdown",
    "id": "c4a6abfd56b8fa1e",

notebooks/getting_started/README.md

@@ -140,7 +140,7 @@ from its component parts. The notebook details how to add VRS identifiers to the
 ### 4 Exploring the Allele Translator
 
 The current implementation of vrs-python facilitates transformation of variants
-in a number of different variant nomenclatures (SPDI, HGVS, gnomAD and Beacon) to VRS form. In the
+in a number of different variant nomenclatures (SPDI, HGVS, and gnomAD) to VRS form. In the
 [Exploring the Allele Translator](4_Exploring_the_AlleleTranslator.ipynb) notebook,
 we show how to transform basic variants to VRS, and in some cases, back to the original nomenclature of the variant.
 

src/ga4gh/vrs/extras/translator.py

@@ -1,6 +1,6 @@
 """Translates various external formats into VRS models.
 
-Input formats: VRS (serialized), hgvs, spdi, gnomad (vcf), beacon
+Input formats: VRS (serialized), hgvs, spdi, gnomad (vcf)
 Output formats: VRS (serialized), hgvs, spdi, gnomad (vcf)
 
 """
@@ -11,8 +11,6 @@
 from collections.abc import Mapping
 from typing import Protocol
 
-from typing_extensions import deprecated
-
 from ga4gh.core import ga4gh_identify
 from ga4gh.vrs import models, normalize
 from ga4gh.vrs.dataproxy import SequenceProxy, _DataProxy
@@ -58,9 +56,6 @@ class _Translator(ABC):  # noqa: B024
 
     """
 
-    beacon_re = re.compile(
-        r"(?P<chr>[^-]+)\s*:\s*(?P<pos>\d+)\s*(?P<ref>\w+)\s*>\s*(?P<alt>\w+)"
-    )
     gnomad_re = re.compile(
         r"(?P<chr>[^-]+)-(?P<pos>\d+)-(?P<ref>[ACGTURYKMSWBDHVN]+)-(?P<alt>[ACGTURYKMSWBDHVN]+)",
         re.IGNORECASE,
@@ -101,7 +96,7 @@ def translate_from(
                     EFO:0030067 for deletions and EFO:0030070 for duplications
             For AlleleTranslator
                 assembly_name (str): Assembly used for `var`. Defaults to the
-                    `default_assembly_name`. Only used for beacon and gnomad.
+                    `default_assembly_name`. Only used for gnomad.
                 require_validation (bool): If `True` then validation checks must pass in
                     order to return a VRS object. A `DataProxyValidationError` will be
                     raised if validation checks fail. If `False` then VRS object will be
@@ -180,7 +175,6 @@ def __init__(
         super().__init__(data_proxy, default_assembly_name, identify)
 
         self.from_translators = {
-            "beacon": self._from_beacon,
             "gnomad": self._from_gnomad,
             "hgvs": self._from_hgvs,
             "spdi": self._from_spdi,
@@ -215,68 +209,6 @@ def _create_allele(self, values: dict, **kwargs) -> models.Allele:
         allele = models.Allele(location=location, state=state)
         return self._post_process_imported_allele(allele, **kwargs)
 
-    @deprecated("This method does not match the Beacon spec and will be removed in v3.")
-    def _from_beacon(self, beacon_expr: str, **kwargs) -> models.Allele | None:
-        """Parse beacon expression into VRS Allele
-
-        kwargs:
-            assembly_name (str): Assembly used for `beacon_expr`.
-            rle_seq_limit Optional(int): If RLE is set as the new state after
-                normalization, this sets the limit for the length of the `sequence`.
-                To exclude `sequence` from the response, set to 0.
-                For no limit, set to `None`.
-                Defaults value set in instance variable, `rle_seq_limit`.
-            do_normalize (bool): `True` if fully justified normalization should be
-                performed. `False` otherwise. Defaults to `True`
-
-        #>>> a = tlr.from_beacon("19 : 44908822 C > T")
-        #>>> a.model_dump()
-        {
-          'location': {
-            'end': 44908822,
-            'start': 44908821,
-            'sequenceReference': {
-              'type': 'SequenceReference',
-              'refgetAccession': 'SQ.IIB53T8CNeJJdUqzn9V_JnRtQadwWCbl'
-            },
-            'type': 'SequenceLocation'
-          },
-          'state': {
-            'sequence': 'C',
-            'type': 'LiteralSequenceExpression'
-          },
-          'type': 'Allele'
-        }
-
-        """
-        if not isinstance(beacon_expr, str):
-            return None
-
-        m = self.beacon_re.match(beacon_expr.replace(" ", ""))
-        if not m:
-            return None
-
-        g = m.groupdict()
-        assembly_name = kwargs.get("assembly_name", self.default_assembly_name)
-        sequence = assembly_name + ":" + g["chr"]
-        refget_accession = self.data_proxy.derive_refget_accession(sequence)
-        if not refget_accession:
-            return None
-
-        start = int(g["pos"]) - 1
-        ref = g["ref"]
-        alt = g["alt"]
-        end = start + len(ref)
-        ins_seq = alt
-
-        values = {
-            "refget_accession": refget_accession,
-            "start": start,
-            "end": end,
-            "literal_sequence": ins_seq,
-        }
-        return self._create_allele(values, **kwargs)
-
     def _from_gnomad(self, gnomad_expr: str, **kwargs) -> models.Allele | None:
         """Parse gnomAD-style VCF expression into VRS Allele
 

tests/extras/cassettes/test_from_beacon.yaml

@@ -17,7 +17,6 @@ def tlr(rest_dataproxy):
 # https://www.ncbi.nlm.nih.gov/clinvar/variation/17848/?new_evidence=true
 snv_inputs = {
     "hgvs": "NC_000019.10:g.44908822C>T",
-    "beacon": "19 : 44908822 C > T",
     "spdi": "NC_000019.10:44908821:1:T",
     "gnomad": "19-44908822-C-T",
 }
@@ -39,7 +38,6 @@ def tlr(rest_dataproxy):
 # https://www.ncbi.nlm.nih.gov/clinvar/variation/693259/?new_evidence=true
 mito_inputs = {
     "hgvs": "NC_012920.1:m.10083A>G",
-    "beacon": "MT : 10083 A > G",
     "spdi": "NC_012920.1:10082:A:G",
     "gnomad": "MT-10083-A-G",
 }
@@ -327,31 +325,11 @@ def test_rle_round_trip_gnomad_spdi(tlr):
 
 def test_from_invalid(tlr):
     with pytest.raises(
-        ValueError, match="Unable to parse data as beacon, gnomad, hgvs, spdi, vrs"
+        ValueError, match="Unable to parse data as gnomad, hgvs, spdi, vrs"
     ):
         tlr.translate_from("BRAF amplication")
 
 
-@pytest.mark.vcr
-def test_from_beacon(tlr):
-    do_normalize = False
-    with pytest.deprecated_call():
-        assert (
-            tlr._from_beacon(
-                snv_inputs["beacon"], do_normalize=do_normalize
-            ).model_dump(exclude_none=True)
-            == snv_output
-        )
-
-    with pytest.deprecated_call():
-        assert (
-            tlr._from_beacon(
-                mito_inputs["beacon"], do_normalize=do_normalize
-            ).model_dump(exclude_none=True)
-            == mito_output
-        )
-
-
 @pytest.mark.vcr
 def test_from_gnomad(tlr):
     do_normalize = False
@@ -977,9 +955,6 @@ def test_normalize_microsatellite_counts(tlr, case):
 # @pytest.mark.vcr
 # def test_errors(tlr):
 #     with pytest.raises(ValueError):
-#         tlr._from_beacon("bogus")
-#
-#     with pytest.raises(ValueError):
 #         tlr._from_gnomad("NM_182763.2:c.688+403C>T")
 #
 #     with pytest.raises(ValueError):