Misc. improvements, unit tests and mypy proofing

.github/workflows/anglerfish.yml

@@ -7,7 +7,7 @@ jobs:
     runs-on: ubuntu-latest
     strategy:
       matrix:
-        python-version: [3.10.2]
+        python-version: [3.12]
 
     steps:
       # Checkout code and install miniconda + environment
@@ -37,10 +37,10 @@ jobs:
       - shell: bash -l {0}
         name: Run anglerfish with test data
         run: |
-          anglerfish run -s test/samples.csv
+          anglerfish run -s testdata/samples.csv
 
       # Run anglerfish explore
       - shell: bash -l {0}
         name: Run anglerfish explore
         run: |
-          anglerfish explore -f test/BC18_P14351_1001.fastq.gz -o test/explore_output
+          anglerfish explore -f testdata/BC18_P14351_1001.fastq.gz -o explore_output

.github/workflows/lint-code.yml

@@ -16,7 +16,7 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v4
         with:
-          python-version: "3.10"
+          python-version: "3.12"
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
@@ -34,7 +34,7 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v4
         with:
-          python-version: "3.10"
+          python-version: "3.12"
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
@@ -51,17 +51,17 @@ jobs:
       - name: Set up Python
         uses: actions/setup-python@v4
         with:
-          python-version: "3.10"
+          python-version: "3.12"
       - name: Install dependencies
         run: |
           python -m pip install --upgrade pip
           pip install mypy
       # Start by installing type stubs
       - name: mypy --> Install stubs
-        run: echo -e "y" | mypy --install-types **/*.py || exit 0
+        run: echo -e "y" | mypy --install-types . || exit 0
       - name: mypy --> Static type checking
         # Configured in pyprojet.toml
-        run: mypy **/*.py
+        run: mypy .
 
     # Use Prettier to check various file formats
   prettier:

.github/workflows/pypi.yml

@@ -11,10 +11,10 @@ jobs:
       - uses: actions/checkout@v4
         name: Check out source-code repository
 
-      - name: Set up Python 3.10
+      - name: Set up Python 3.12
         uses: actions/setup-python@v4
         with:
-          python-version: 3.10.10
+          python-version: 3.12
 
       - name: Install python dependencies
         run: |

.github/workflows/test-code.yml

@@ -0,0 +1,44 @@
+name: test-code
+on: [push, pull_request]
+
+# Cancel if a newer run is started
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
+jobs:
+  run_pytest:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: [3.12]
+
+    steps:
+      # Checkout code and install miniconda + environment
+      - uses: actions/checkout@v4
+      - uses: mamba-org/setup-micromamba@v1
+        with:
+          init-shell: bash
+          create-args: >-
+            python=${{ matrix.python-version }}
+            pip
+          environment-file: environment.yml
+          environment-name: anglerfish-dev
+
+      # Install Anglerfish
+      - shell: bash -l {0}
+        name: Install Anglerfish
+        run: |
+          python -m pip install .
+
+      # Install Pytest
+      - shell: bash -l {0}
+        name: Install Pytest
+        run: |
+          python -m pip install pytest
+
+      # Run Pytest
+      - shell: bash -l {0}
+        name: Run Pytest
+        run: |
+          pytest .

.gitignore

@@ -1,9 +1,11 @@
+*.egg-info
 *.pyc
 *~
-*.egg-info
+.*_cache
 .DS_Store
 .benchmarks
-.*_cache
-node_modules
+.ignoredir
 .vscode
+__pycache__
 build
+node_modules

anglerfish/anglerfish.py

@@ -13,6 +13,7 @@
 import numpy as np
 import pkg_resources
 
+from .demux.adaptor import Adaptor
 from .demux.demux import (
     cluster_matches,
     layout_matches,
@@ -86,14 +87,16 @@ def run_demux(args):
     adaptor_set: set[tuple[str, str]] = set(adaptor_tuples)
 
     # Create a dictionary with the adaptors as keys and an empty list as value
-    adaptors_sorted: dict[tuple[str, str], list] = dict([(i, []) for i in adaptor_set])
+    adaptors_sorted: dict[tuple[str, str], list[tuple[str, Adaptor, str]]] = dict(
+        [(i, []) for i in adaptor_set]
+    )
 
     # Populate the dictionary values with sample-specific information
     """
         adaptors_sorted = {
-            ( adaptor_name, ont_barcode ) : [
-                (sample_name, adaptor, fastq),
-                (sample_name, adaptor, fastq),
+             adaptor_name_str, ont_barcode_str ) : [
+                (sample_name_str, Adaptor, fastq_str),
+                (sample_name_str, Adaptor, fastq_str),
                 ...
             ],
             ...
@@ -168,7 +171,7 @@ def run_demux(args):
                 **flips[args.force_rc],
             )
             flipped_i7, flipped_i5 = flips[args.force_rc].values()
-        elif args.lenient:  # Try reverse complementing the I5 and/or i7 indices and choose the best match
+        elif args.lenient:  # Try reverse complementing the i5 and/or i7 indices and choose the best match
             flipped = {}
             results = []
             pool = multiprocessing.Pool(

anglerfish/demux/adaptor.py

@@ -63,10 +63,36 @@ class AdaptorPart:
     """This class is used for the i5 or i7 adaptor."""
 
     def __init__(self, sequence_token: str, name: str, index_seq: str | None):
+        ## Type declaration of attributes to be assigned upon instantiation
+        # Attributes from arguments
+        self.name: str
+        self.sequence_token: str
+        self.index_seq: str | None
+
+        # Index attributes
+        self.has_index: bool
+        self.len_index: int | None
+        self.len_before_index: int | None
+        self.len_after_index: int | None
+
+        # UMI attributes
+        self.has_umi: bool
+        self.len_umi: int | None
+        self.len_umi_before_index: int | None
+        self.len_umi_after_index: int | None
+
+        # Length attributes
+        self.len_total: int | None
+        self.len_constant: int
+
+        # Instantiation outsorced to private method
+        self._setup(sequence_token, name, index_seq)
+
+    def _setup(self, sequence_token: str, name: str, index_seq: str | None):
         # Assign attributes from args
-        self.sequence_token: str = sequence_token
-        self.name: str = name
-        self.index_seq: str | None = index_seq
+        self.sequence_token = sequence_token
+        self.name = name
+        self.index_seq = index_seq
 
         # Index bool and len
         if has_match(INDEX_TOKEN, self.sequence_token):
@@ -76,6 +102,10 @@ def __init__(self, sequence_token: str, name: str, index_seq: str | None):
             self.len_index = len(index_seq) if index_seq else None
 
         else:
+            if self.index_seq is not None:
+                raise UserWarning(
+                    "Index sequence specified, but no index token found in adaptor sequence."
+                )
             self.has_index = False
             self.len_index = 0
 
@@ -87,21 +117,13 @@ def __init__(self, sequence_token: str, name: str, index_seq: str | None):
             )
         elif len(umi_tokens) == 1:
             self.has_umi = True
-            self.len_umi = int(
-                re.search(UMI_LENGTH_TOKEN, self.sequence_token).group(1)
-            )
+            umi_token_search = re.search(UMI_LENGTH_TOKEN, self.sequence_token)
+            assert isinstance(umi_token_search, re.Match)
+            self.len_umi = int(umi_token_search.group(1))
         else:
             self.has_umi = False
             self.len_umi = 0
 
-        # Type declaration of attributes to be assigned
-        self.len_before_index: int | None
-        self.len_after_index: int | None
-        self.len_umi_before_index: int | None
-        self.len_umi_after_index: int | None
-        self.len_total: int | None
-        self.len_constant: int
-
         # Lengths
         if self.has_index and self.has_umi:
             # Index and UMI
@@ -149,7 +171,12 @@ def __init__(self, sequence_token: str, name: str, index_seq: str | None):
             self.len_before_index = None
             self.len_after_index = None
 
-        self.len_total = len(self.get_mask(insert_Ns=True)) if self.index_seq else None
+        if (
+            self.has_index is True and self.index_seq is not None
+        ) or self.has_index is False:
+            self.len_total = len(self.get_mask(insert_Ns=True))
+        else:
+            self.len_total = None
         self.len_constant = len(self.get_mask(insert_Ns=False))
 
     def get_mask(self, insert_Ns: bool = True) -> str:
@@ -165,11 +192,12 @@ def get_mask(self, insert_Ns: bool = True) -> str:
             else 0
         )
 
-        umi_mask_length = (
-            max(self.len_umi_after_index, self.len_umi_before_index)
-            if insert_Ns and self.has_umi
-            else 0
-        )
+        if insert_Ns and self.has_umi:
+            assert self.len_umi_before_index is not None
+            assert self.len_umi_after_index is not None
+            umi_mask_length = max(self.len_umi_after_index, self.len_umi_before_index)
+        else:
+            umi_mask_length = 0
 
         # Test if the index is specified in the adaptor sequence when it shouldn't be
         if (
@@ -189,7 +217,7 @@ def get_mask(self, insert_Ns: bool = True) -> str:
             return self.sequence_token
 
 
-def has_match(pattern: re.Pattern, query: str) -> bool:
+def has_match(pattern: re.Pattern | str, query: str) -> bool:
     """General function to check if a string contains a pattern."""
     match = re.search(pattern, query)
     if match is None:
@@ -209,6 +237,8 @@ def validate_adaptors(adaptors_dict: dict):
                     f"Adaptor {adaptor_name} has an invalid sequence for {i}: {sequence_token}. Does not conform to the pattern {VALID_SEQUENCE_TOKEN_PATTERN}."
                 )
 
+    return True
+
 
 def load_adaptors(raw: bool = False) -> list[Adaptor] | dict:
     """Fetch all adaptors.
@@ -226,7 +256,7 @@ def load_adaptors(raw: bool = False) -> list[Adaptor] | dict:
         adaptors_dict = yaml.safe_load(f)
 
     # Validate input
-    validate_adaptors(adaptors_dict)
+    assert validate_adaptors(adaptors_dict) is True
 
     # Optionally, return raw dict
     if raw: