Fixed all existing tests and removed some seemingly unnecessary "config" files.

To do: add new tests to cover all the new code, and convert the
existing config CSV files (which are intentionally an old version,
and used in bblab) to our YAML format.

Author: rhliang

src/easyhla/bblab.py

@@ -88,7 +88,7 @@ def process_from_file_to_files(
 
     log_and_print(
         f"Run commencing {time_start.strftime(DATE_FORMAT)}. "
-        f"Allele definitions last updated {hla_alg.last_modified}.",
+        f"Allele definitions last updated {hla_alg.last_updated}.",
         to_stdout=to_stdout,
     )
 

src/easyhla/default_data/hla_a_seq.csv

@@ -31,7 +31,7 @@
 
 class LoadedStandards(TypedDict):
     tag: str
-    last_modified: datetime
+    last_updated: datetime
     standards: dict[HLA_LOCUS, dict[str, HLAStandard]]
 
 
@@ -63,7 +63,7 @@ def __init__(
         self.hla_standards: dict[HLA_LOCUS, dict[str, HLAStandard]] = loaded_standards[
             "standards"
         ]
-        self.last_modified: datetime = loaded_standards["last_modified"]
+        self.last_updated: datetime = loaded_standards["last_updated"]
         self.tag: str = loaded_standards["tag"]
 
         self.hla_frequencies: dict[HLA_LOCUS, dict[HLAProteinPair, int]]
@@ -121,7 +121,7 @@ def read_hla_standards(standards_io: TextIOBase) -> LoadedStandards:
 
         return {
             "tag": stored_stds.tag,
-            "last_modified": stored_stds.last_modified,
+            "last_updated": stored_stds.last_updated,
             "standards": hla_stds,
         }
 
@@ -161,8 +161,8 @@ def read_hla_frequencies(
             "B": {},
             "C": {},
         }
-        for locus in ("A", "B", "C"):
-            for line in frequencies_io.readlines():
+        for line in frequencies_io.readlines():
+            for locus in ("A", "B", "C"):
                 column_id = EasyHLA.COLUMN_IDS[locus]
                 line_array = line.strip().split(",")[column_id : column_id + 2]
 
@@ -429,7 +429,7 @@ def interpret(
         b5701_standards: Optional[list[HLAStandard]] = None
         if locus == "B":
             b5701_standards = [
-                self.hla_standards[allele] for allele in self.B5701_ALLELES
+                self.hla_standards[locus][allele] for allele in self.B5701_ALLELES
             ]
 
         return HLAInterpretation(
@@ -440,10 +440,11 @@ def interpret(
                     mismatches=self.get_mismatches(
                         combined_std.standard_bin,
                         seq,
+                        locus,
                     ),
                 )
                 for combined_std, mismatch_count in all_combos.items()
             },
-            allele_frequencies=self.hla_frequencies,
+            allele_frequencies=self.hla_frequencies[locus],
             b5701_standards=b5701_standards,
         )

src/easyhla/default_data/hla_b_seq.csv

@@ -1,12 +1,19 @@
 import re
 from collections.abc import Iterable
 from operator import itemgetter
-from typing import Optional
+from typing import Optional, Self
 
 import numpy as np
 from pydantic import BaseModel, ConfigDict
 
-from .utils import HLA_LOCUS, allele_coordinates, bin2nuc, count_forgiving_mismatches
+from .utils import (
+    HLA_LOCUS,
+    HLARawStandard,
+    allele_coordinates,
+    bin2nuc,
+    count_forgiving_mismatches,
+    nuc2bin,
+)
 
 
 class HLASequence(BaseModel):
@@ -52,6 +59,14 @@ def sequence(self) -> tuple[int, ...]:
     def sequence_np(self) -> np.ndarray:
         return np.array(self.sequence)
 
+    @classmethod
+    def from_raw_standard(cls, raw_standard: HLARawStandard) -> Self:
+        return cls(
+            allele=raw_standard.allele,
+            two=nuc2bin(raw_standard.exon2),
+            three=nuc2bin(raw_standard.exon3),
+        )
+
 
 class HLAStandardMatch(HLAStandard):
     mismatch: int

src/easyhla/default_data/hla_c_seq.csv

@@ -4,7 +4,6 @@
 from collections import defaultdict
 from collections.abc import Iterable, Sequence
 from datetime import datetime
-from operator import attrgetter
 from typing import Final, Literal, Optional, Self
 
 import numpy as np
@@ -487,8 +486,7 @@ def group_identical_alleles(
             logger.info(
                 f"[{', '.join(grouped_allele.alleles)}] -> {grouped_allele.name}"
             )
-
-    return sorted(grouped_alleles, key=attrgetter("name"))
+    return grouped_alleles
 
 
 def compute_stored_standard_checksum(
@@ -532,3 +530,4 @@ def compute_compare_checksum(self) -> Self:
         else:
             if self.checksum != checksum:
                 raise ValueError("Checksum mismatch")
+        return self

src/easyhla/default_data/hla_standards.yaml

@@ -17,10 +17,10 @@
     HLASequence,
     HLAStandard,
 )
-from easyhla.utils import EXON_NAME, HLA_LOCUS, nuc2bin
+from easyhla.utils import EXON_NAME, HLA_LOCUS, HLARawStandard, nuc2bin
 
 from .clinical_hla_lib_test import DUMMY_FREQUENCIES, dummy_hla_sequence, dummy_matches
-from .easyhla_test import HLA_STANDARDS, DummyStandard
+from .easyhla_test import HLA_STANDARDS
 
 
 @pytest.mark.parametrize(
@@ -732,7 +732,7 @@ def test_pair_exons(
     paired_seqs: list[HLASequence]
     unmatched: dict[EXON_NAME, dict[str, Seq]]
 
-    current_standard: DummyStandard = HLA_STANDARDS[locus]
+    current_standard: HLARawStandard = HLA_STANDARDS[locus]
     fake_standard: HLAStandard = HLAStandard(
         allele=current_standard.allele,
         two=nuc2bin(current_standard.exon2),

src/easyhla/easyhla.py

@@ -1,6 +1,6 @@
 from typing import Tuple
 
-from easyhla import EasyHLA
+from easyhla.easyhla import EasyHLA
 
 
 def make_comparison(easyhla: EasyHLA, ref_seq: str, test_seq: str) -> str: