feat: k-mer-seeded top-k ungapped local alignment (0.6.0)

.github/workflows/lint.yaml

@@ -15,6 +15,13 @@ jobs:
       with:
         python-version: '3.11'
 
+    - name: Install Rust toolchain
+      uses: actions-rs/toolchain@v1
+      with:
+        toolchain: stable
+        override: true
+        components: rustfmt
+
     - name: Install packages
       run: pip install -r requirements-dev.txt
 

.pre-commit-config.yaml

@@ -33,6 +33,14 @@ repos:
         args: ["--fix"]
       - id: ruff-format
 
+  - repo: local
+    hooks:
+      - id: rustfmt
+        name: rustfmt
+        entry: rustfmt --check --edition 2021
+        language: system
+        types: [rust]
+
   - repo: https://github.com/pre-commit/mirrors-mypy
     rev: v1.18.2
     hooks:

Cargo.toml

@@ -1,6 +1,6 @@
 [package]
 name = "seq_smith"
-version = "0.5.1"
+version = "0.6.0"
 edition = "2021"
 
 [dependencies]

build.rs

@@ -1,3 +1,3 @@
 fn main() {
-  pyo3_build_config::add_extension_module_link_args();
+    pyo3_build_config::add_extension_module_link_args();
 }

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "maturin"
 
 [project]
 name = "seq-smith"
-version = "0.5.1"
+version = "0.6.0"
 authors = [
     { name = "Tobias Sargeant", email = "tobias.sargeant@gmail.com" },
 ]

seq_smith/__init__.py

@@ -11,6 +11,7 @@
     overlap_align,
     overlap_align_many,
     top_k_ungapped_local_align,
+    top_k_ungapped_local_align_kmer,
     top_k_ungapped_local_align_many,
 )
 from .python_utils import decode, encode, format_alignment_ascii, generate_cigar, make_score_matrix
@@ -33,5 +34,6 @@
     "overlap_align",
     "overlap_align_many",
     "top_k_ungapped_local_align",
+    "top_k_ungapped_local_align_kmer",
     "top_k_ungapped_local_align_many",
 ]

seq_smith/_seq_smith.pyi

@@ -125,6 +125,18 @@ def top_k_ungapped_local_align(
     filter_overlap_a: bool = True,
     filter_overlap_b: bool = True,
 ) -> list[Alignment]: ...
+def top_k_ungapped_local_align_kmer(
+    seqa: bytes,
+    seqb: bytes,
+    score_matrix: npt.NDArray[np.int32],
+    k: int,
+    kmer_size: int,
+    max_hits_per_kmer: int,
+    max_hit_gap: int = 20,
+    min_kmer_hits_per_span: int = 1,
+    filter_overlap_a: bool = True,
+    filter_overlap_b: bool = True,
+) -> list[Alignment]: ...
 def top_k_ungapped_local_align_many(
     seqa: bytes,
     seqbs: Sequence[bytes],