Added predetermined headers feature

.vscode/extensions.json

@@ -1,5 +1,6 @@
 {
     "recommendations": [
-        "piotrpalarz.vscode-gitignore-generator"
+        "piotrpalarz.vscode-gitignore-generator",
+        "gruntfuggly.todo-tree"
     ]
 }

Jenkinsfile

@@ -36,7 +36,9 @@ pipeline {
             parallel {
                 stage ("git.reslate.systems") {
                     when {
-                        branch '**/main'
+                        not {
+                            tag '*.*.*'
+                        }
                     }
 
                     environment {
@@ -49,7 +51,7 @@ pipeline {
                 }
                 stage ("pypi.org") {
                     when {
-                        tag '*.*'
+                        tag '*.*.*'
                     }
                     environment {
                         TOKEN = credentials('pypi.org')

autobigs-engine/meta.yaml

@@ -1,44 +0,0 @@
-{% set name = "autoBIGS.engine" %}
-{% set version = "0.12.1.dev1+gb8cebb8.d20250221" %}
-
-package:
-  name: {{ name|lower|replace(".", "-") }}
-  version: {{ version }}
-
-source:
-  url: file:///workspaces/autoBIGS.engine/dist/autobigs_engine-0.12.1.dev1%2Bgb8cebb8.d20250221.tar.gz
-  sha256: c86441b94f935cfa414ff28ca4c026a070e0fb15988ea3bb7d1a942859a09b16
-
-build:
-  noarch: python
-  script: {{ PYTHON }} -m pip install . -vv --no-deps --no-build-isolation
-  number: 0
-  run_exports:
-    - {{ pin_subpackage( name|lower|replace(".", "-"), max_pin="x.x") }}
-requirements:
-  host:
-    - python >=3.12
-    - setuptools >=64
-    - setuptools-scm >=8
-    - pip
-  run:
-    - python >=3.12
-    - biopython ==1.85
-    - aiohttp ==3.11.*
-
-test:
-  imports:
-    - autobigs
-  commands:
-    - pip check
-  requires:
-    - pip
-
-about:
-  summary: A library to rapidly fetch fetch MLST profiles given sequences for various diseases.
-  license: GPL-3.0-or-later
-  license_file: LICENSE
-  home: https://github.com/Syph-and-VPD-Lab/autoBIGS.engine
-extra:
-  recipe-maintainers:
-    - Harrison Deng

src/autobigs/engine/analysis/bigsdb.py

@@ -9,7 +9,7 @@ import shutil
 import tempfile
 from typing import Any, AsyncGenerator, AsyncIterable, Coroutine, Iterable, Mapping, Sequence, Set, Union
 
-from aiohttp import ClientSession, ClientTimeout
+from aiohttp import ClientOSError, ClientSession, ClientTimeout, ServerDisconnectedError
 
 from autobigs.engine.reading import read_fasta
 from autobigs.engine.structures.alignment import PairwiseAlignment
@@ -43,11 +43,12 @@ class BIGSdbMLSTProfiler(AbstractAsyncContextManager):
 
 class RemoteBIGSdbMLSTProfiler(BIGSdbMLSTProfiler):
 
-    def __init__(self, database_api: str, database_name: str, scheme_id: int):
+    def __init__(self, database_api: str, database_name: str, scheme_id: int, retry_requests: int = 5):
+        self._retry_limit = retry_requests
         self._database_name = database_name
         self._scheme_id = scheme_id
         self._base_url = f"{database_api}/db/{self._database_name}/schemes/{self._scheme_id}/"
-        self._http_client = ClientSession(self._base_url, timeout=ClientTimeout(60))
+        self._http_client = ClientSession(self._base_url, timeout=ClientTimeout(300))
 
     async def __aenter__(self):
         return self
@@ -57,40 +58,59 @@ class RemoteBIGSdbMLSTProfiler(BIGSdbMLSTProfiler):
         uri_path = "sequence"
         if isinstance(query_sequence_strings, str) or isinstance(query_sequence_strings, NamedString):
             query_sequence_strings = [query_sequence_strings]
+        
         for sequence_string in query_sequence_strings:
-            async with self._http_client.post(uri_path, json={
-                "sequence": sequence_string if isinstance(sequence_string, str) else sequence_string.sequence,
-                "partial_matches": True
-            }) as response:
-                sequence_response: dict = await response.json()
+            attempts = 0
+            success = False
+            last_error = None
+            while not success and attempts < self._retry_limit:
+                attempts += 1
+                request = self._http_client.post(uri_path, json={
+                    "sequence": sequence_string if isinstance(sequence_string, str) else sequence_string.sequence,
+                    "partial_matches": True
+                })
+                try:
+                    async with request as response:
+                        sequence_response: dict = await response.json()
 
-                if "exact_matches" in sequence_response:
-                    # loci -> list of alleles with id and loci
-                    exact_matches: dict[str, Sequence[dict[str, str]]] = sequence_response["exact_matches"]  
-                    for allele_loci, alleles in exact_matches.items():
-                        for allele in alleles:
-                            alelle_id = allele["allele_id"]
-                            result_allele = Allele(allele_locus=allele_loci, allele_variant=alelle_id, partial_match_profile=None)
-                            yield result_allele if isinstance(sequence_string, str) else (sequence_string.name, result_allele)
-                elif "partial_matches" in sequence_response:
-                    partial_matches: dict[str, dict[str, Union[str, float, int]]] = sequence_response["partial_matches"] 
-                    for allele_loci, partial_match in partial_matches.items():
-                        if len(partial_match) <= 0:
-                            continue
-                        partial_match_profile = AlignmentStats(
-                            percent_identity=float(partial_match["identity"]),
-                            mismatches=int(partial_match["mismatches"]),
-                            gaps=int(partial_match["gaps"]),
-                            match_metric=int(partial_match["bitscore"])
-                        )
-                        result_allele = Allele(
-                            allele_locus=allele_loci,
-                            allele_variant=str(partial_match["allele"]),
-                            partial_match_profile=partial_match_profile
-                        )
-                        yield result_allele if isinstance(sequence_string, str) else (sequence_string.name, result_allele)
+                        if "exact_matches" in sequence_response:
+                            # loci -> list of alleles with id and loci
+                            exact_matches: dict[str, Sequence[dict[str, str]]] = sequence_response["exact_matches"]  
+                            for allele_loci, alleles in exact_matches.items():
+                                for allele in alleles:
+                                    alelle_id = allele["allele_id"]
+                                    result_allele = Allele(allele_locus=allele_loci, allele_variant=alelle_id, partial_match_profile=None)
+                                    yield result_allele if isinstance(sequence_string, str) else (sequence_string.name, result_allele)
+                        elif "partial_matches" in sequence_response:
+                            partial_matches: dict[str, dict[str, Union[str, float, int]]] = sequence_response["partial_matches"] 
+                            for allele_loci, partial_match in partial_matches.items():
+                                if len(partial_match) <= 0:
+                                    continue
+                                partial_match_profile = AlignmentStats(
+                                    percent_identity=float(partial_match["identity"]),
+                                    mismatches=int(partial_match["mismatches"]),
+                                    gaps=int(partial_match["gaps"]),
+                                    match_metric=int(partial_match["bitscore"])
+                                )
+                                result_allele = Allele(
+                                    allele_locus=allele_loci,
+                                    allele_variant=str(partial_match["allele"]),
+                                    partial_match_profile=partial_match_profile
+                                )
+                                yield result_allele if isinstance(sequence_string, str) else (sequence_string.name, result_allele)
+                        else:
+                            raise NoBIGSdbMatchesException(self._database_name, self._scheme_id, sequence_string.name if isinstance(sequence_string, NamedString) else None)
+                except (ConnectionError, ServerDisconnectedError, ClientOSError) as e: # Errors we will retry
+                    last_error = e
+                    success = False
+                    await asyncio.sleep(5) # In case the connection issue is due to rate issues
                 else:
-                    raise NoBIGSdbMatchesException(self._database_name, self._scheme_id, sequence_string.name if isinstance(sequence_string, NamedString) else None)
+                    success = True
+            if not success and last_error is not None:
+                try:
+                    raise last_error
+                except (ConnectionError, ServerDisconnectedError, ClientOSError) as e: # Non-fatal errors
+                    yield Allele("error", "error", None)
 
     async def determine_mlst_st(self, alleles: Union[AsyncIterable[Union[Allele, tuple[str, Allele]]], Iterable[Union[Allele, tuple[str, Allele]]]]) -> Union[MLSTProfile, NamedMLSTProfile]:
         uri_path = "designations"
@@ -113,23 +133,43 @@ class RemoteBIGSdbMLSTProfiler(BIGSdbMLSTProfiler):
         request_json = {
             "designations": allele_request_dict
         }
-        async with self._http_client.post(uri_path, json=request_json) as response:
-            response_json: dict = await response.json()
-            allele_set: Set[Allele] = set()
-            response_json.setdefault("fields", dict())
-            scheme_fields_returned: dict[str, str] = response_json["fields"]
-            scheme_fields_returned.setdefault("ST", "unknown")
-            scheme_fields_returned.setdefault("clonal_complex", "unknown")
-            scheme_exact_matches: dict = response_json["exact_matches"]
-            for exact_match_locus, exact_match_alleles in scheme_exact_matches.items():
-                allele_set.add(Allele(exact_match_locus, exact_match_alleles[0]["allele_id"], None))
-            if len(allele_set) == 0:
-                raise ValueError("Passed in no alleles.")
-            result_mlst_profile = MLSTProfile(allele_set, scheme_fields_returned["ST"], scheme_fields_returned["clonal_complex"])
-            if len(names_list) > 0:
-                result_mlst_profile = NamedMLSTProfile(str(tuple(names_list)) if len(set(names_list)) > 1 else names_list[0], result_mlst_profile)
-            return result_mlst_profile
 
+        attempts = 0
+        success = False
+        last_error = None
+        while attempts < self._retry_limit and not success:
+            attempts += 1
+            try:
+                async with self._http_client.post(uri_path, json=request_json) as response:
+                    response_json: dict = await response.json()
+                    allele_set: Set[Allele] = set()
+                    response_json.setdefault("fields", dict())
+                    scheme_fields_returned: dict[str, str] = response_json["fields"]
+                    scheme_fields_returned.setdefault("ST", "unknown")
+                    scheme_fields_returned.setdefault("clonal_complex", "unknown")
+                    scheme_exact_matches: dict = response_json["exact_matches"]
+                    for exact_match_locus, exact_match_alleles in scheme_exact_matches.items():
+                        allele_set.add(Allele(exact_match_locus, exact_match_alleles[0]["allele_id"], None))
+                    if len(allele_set) == 0:
+                        raise ValueError("Passed in no alleles.")
+                    result_mlst_profile = MLSTProfile(allele_set, scheme_fields_returned["ST"], scheme_fields_returned["clonal_complex"])
+                    if len(names_list) > 0:
+                        result_mlst_profile = NamedMLSTProfile(str(tuple(names_list)) if len(set(names_list)) > 1 else names_list[0], result_mlst_profile)
+                    return result_mlst_profile
+            except (ConnectionError, ServerDisconnectedError, ClientOSError) as e:
+                last_error = e
+                success = False
+                await asyncio.sleep(5)
+            else:
+                success = True
+        try:
+            if last_error is not None:
+                raise last_error
+        except (ConnectionError, ServerDisconnectedError, ClientOSError) as e:
+                result_mlst_profile = NamedMLSTProfile((str(tuple(names_list)) if len(set(names_list)) > 1 else names_list[0]) + ":Error", None)
+        raise ValueError("Last error was not recorded.")
+
+                    
     async def profile_string(self, query_sequence_strings: Iterable[Union[NamedString, str]]) -> Union[NamedMLSTProfile, MLSTProfile]:
         alleles = self.determine_mlst_allele_variants(query_sequence_strings)
         return await self.determine_mlst_st(alleles)
@@ -211,6 +251,16 @@ class BIGSdbIndex(AbstractAsyncContextManager):
 
     async def build_profiler_from_seqdefdb(self, local: bool, dbseqdef_name: str, scheme_id: int) -> BIGSdbMLSTProfiler:
         return get_BIGSdb_MLST_profiler(local, await self.get_bigsdb_api_from_seqdefdb(dbseqdef_name), dbseqdef_name, scheme_id)
+    
+    async def get_scheme_loci(self, dbseqdef_name: str, scheme_id: int) -> list[str]:
+        uri_path = f"{await self.get_bigsdb_api_from_seqdefdb(dbseqdef_name)}/db/{dbseqdef_name}/schemes/{scheme_id}"
+        async with self._http_client.get(uri_path) as response:
+            response_json = await response.json()
+            loci = response_json["loci"]
+            results = []
+            for locus in loci:
+                results.append(path.basename(locus))
+            return results
 
     async def close(self):
         await self._http_client.close()

src/autobigs/engine/reading.py

@@ -1,5 +1,6 @@
 import asyncio
 from io import TextIOWrapper
+from os import path
 from typing import Any, AsyncGenerator, Iterable, Union
 from Bio import SeqIO
 
@@ -9,7 +10,7 @@ async def read_fasta(handle: Union[str, TextIOWrapper]) -> Iterable[NamedString]
     fasta_sequences = asyncio.to_thread(SeqIO.parse, handle=handle, format="fasta")
     results = []
     for fasta_sequence in await fasta_sequences:
-        results.append(NamedString(fasta_sequence.id, str(fasta_sequence.seq)))
+        results.append(NamedString("{0}:{1}".format(path.basename(handle.name if isinstance(handle, TextIOWrapper) else handle), fasta_sequence.id), str(fasta_sequence.seq)))
     return results
 
 async def read_multiple_fastas(handles: Iterable[Union[str, TextIOWrapper]]) -> AsyncGenerator[Iterable[NamedString], Any]:

src/autobigs/engine/writing.py

@@ -1,7 +1,7 @@
 from collections import defaultdict
 import csv
 from os import PathLike
-from typing import AsyncIterable, Collection, Mapping, Sequence, Union
+from typing import AsyncIterable, Collection, Iterable, Mapping, Sequence, Union
 
 from autobigs.engine.structures.mlst import Allele, MLSTProfile, NamedMLSTProfile
 
@@ -17,7 +17,7 @@ def alleles_to_text_map(alleles: Collection[Allele]) -> Mapping[str, Union[Seque
             result[locus] = tuple(result[locus]) # type: ignore
     return dict(result)
 
-async def write_mlst_profiles_as_csv(mlst_profiles_iterable: AsyncIterable[NamedMLSTProfile], handle: Union[str, bytes, PathLike[str], PathLike[bytes]]) -> Sequence[str]:
+async def write_mlst_profiles_as_csv(mlst_profiles_iterable: AsyncIterable[NamedMLSTProfile], handle: Union[str, bytes, PathLike[str], PathLike[bytes]], allele_names: Iterable[str]) -> Sequence[str]:
     failed = list()
     with open(handle, "w", newline='') as filehandle:
         header = None
@@ -30,7 +30,7 @@ async def write_mlst_profiles_as_csv(mlst_profiles_iterable: AsyncIterable[Named
                 continue
             allele_mapping = alleles_to_text_map(mlst_profile.alleles)
             if writer is None:
-                header = ["id", "st", "clonal-complex", *sorted(allele_mapping.keys())]
+                header = ["id", "st", "clonal-complex", *sorted(allele_names)]
                 writer = csv.DictWriter(filehandle, fieldnames=header)
                 writer.writeheader()
             row_dictionary = {

tests/autobigs/engine/analysis/test_bigsdb.py

@@ -222,3 +222,12 @@ class TestBIGSdbIndex:
                 assert isinstance(profile, MLSTProfile)
                 assert profile.clonal_complex == "ST-2 complex"
                 assert profile.sequence_type == "1"
+
+
+    @pytest.mark.parametrize(["bigsdb_name", "scheme_id", "expected"], [
+        ("pubmlst_bordetella_seqdef", 3, ["adk", "fumC", "glyA", "tyrB", "icd", "pepA", "pgm"])
+    ])
+    async def test_bigsdb_index_fetches_loci_names(self, bigsdb_name, scheme_id, expected):
+        async with BIGSdbIndex() as bigsdb_index:
+            loci = await bigsdb_index.get_scheme_loci(bigsdb_name, scheme_id)
+            assert set(loci) == set(expected)

tests/autobigs/engine/test_reading.py

@@ -4,4 +4,9 @@ from autobigs.engine.reading import read_fasta
 async def test_fasta_reader_not_none():
     named_strings = await read_fasta("tests/resources/tohama_I_bpertussis.fasta")
     for named_string in named_strings:
-        assert named_string.name == "BX470248.1"
+        assert named_string.name is not None
+
+async def test_fasta_reader_name_contains_file_and_id():
+    named_strings = await read_fasta("tests/resources/tohama_I_bpertussis.fasta")
+    for named_string in named_strings:
+        assert named_string.name == "tohama_I_bpertussis.fasta:BX470248.1"

tests/autobigs/engine/test_writing.py

@@ -27,7 +27,7 @@ async def test_column_order_is_same_as_expected_file(dummy_alphabet_mlst_profile
     dummy_profiles = [dummy_alphabet_mlst_profile]
     with tempfile.TemporaryDirectory() as temp_dir:
         output_path = path.join(temp_dir, "out.csv")
-        await write_mlst_profiles_as_csv(iterable_to_asynciterable(dummy_profiles), output_path)
+        await write_mlst_profiles_as_csv(iterable_to_asynciterable(dummy_profiles), output_path, ["A", "D", "B", "C"])
         with open(output_path) as csv_handle:
             csv_reader = reader(csv_handle)
             lines = list(csv_reader)
@@ -38,7 +38,7 @@ async def test_csv_writing_sample_name_not_repeated_when_single_sequence(dummy_a
     dummy_profiles = [dummy_alphabet_mlst_profile]
     with tempfile.TemporaryDirectory() as temp_dir:
         output_path = path.join(temp_dir, "out.csv")
-        await write_mlst_profiles_as_csv(iterable_to_asynciterable(dummy_profiles), output_path)
+        await write_mlst_profiles_as_csv(iterable_to_asynciterable(dummy_profiles), output_path, ["A", "D", "B", "C"])
         with open(output_path) as csv_handle:
             csv_reader = reader(csv_handle)
             lines = list(csv_reader)
@@ -63,7 +63,7 @@ async def test_csv_writing_includes_asterisk_for_non_exact(dummy_alphabet_mlst_p
     dummy_profiles = [dummy_alphabet_mlst_profile]
     with tempfile.TemporaryDirectory() as temp_dir:
         output_path = path.join(temp_dir, "out.csv")
-        await write_mlst_profiles_as_csv(iterable_to_asynciterable(dummy_profiles), output_path)
+        await write_mlst_profiles_as_csv(iterable_to_asynciterable(dummy_profiles), output_path, ["A", "D", "B", "C"])
         with open(output_path) as csv_handle:
             csv_reader = reader(csv_handle)
             lines = list(csv_reader)