update: loop rep, bootstrap, loop equivalence test

mise.toml

@@ -61,5 +61,10 @@ run = [
     "git push"
 ]
 
+[tasks.lint]
+run = [
+    "uv run ty check",
+]
+
 [tasks.kernel]
 run = "uv run ipython kernel install --name 'scLoop' --user"

pyproject.toml

@@ -106,3 +106,8 @@ executionEnvironments = [
 
 venvPath = "/home/stanfish/Git/scloop"
 venv = ".venv"
+
+[dependency-groups]
+dev = [
+    "ty>=0.0.1a34",
+]

src/scloop/computing/homology.py

@@ -1,15 +1,21 @@
 # Copyright 2025 Zhiyuan Yu (Heemskerk's lab, University of Michigan)
 from __future__ import annotations
 
+from typing import TYPE_CHECKING
+
 import numpy as np
 from anndata import AnnData
 from scipy.sparse import csr_matrix
 from sklearn.neighbors import radius_neighbors_graph
 
 from ..data.metadata import ScloopMeta
-from ..data.ripser_lib import get_boundary_matrix, ripser
+from ..data.ripser_lib import get_boundary_matrix, ripser  # type: ignore[import-not-found]
 from ..data.types import Diameter_t, IndexListDistMatrix
 from ..data.utils import encode_triangles_and_edges
+from ..utils.linear_algebra_gf2.m4ri_lib import solve_multiple_gf2  # type: ignore[import-not-found]
+
+if TYPE_CHECKING:
+    from ..data.containers import BoundaryMatrixD1
 
 
 def compute_sparse_pairwise_distance(
@@ -20,6 +26,8 @@ def compute_sparse_pairwise_distance(
     thresh: Diameter_t | None = None,
     **nei_kwargs,
 ) -> tuple[csr_matrix, IndexListDistMatrix | None]:
+    # important, default is binary graph
+    nei_kwargs.setdefault("mode", "distance")
     assert meta.preprocess is not None
     assert meta.preprocess.embedding_method is not None
     emb = adata.obsm[f"X_{meta.preprocess.embedding_method}"]
@@ -53,10 +61,16 @@ def compute_persistence_diagram_and_cocycles(
     meta: ScloopMeta,
     thresh: Diameter_t | None = None,
     bootstrap: bool = False,
+    noise_scale: float = 1e3,
     **nei_kwargs,
-) -> tuple[list[np.ndarray], list, IndexListDistMatrix | None, csr_matrix]:
+) -> tuple[list, list, IndexListDistMatrix | None, csr_matrix]:
     sparse_pairwise_distance_matrix, boot_idx = compute_sparse_pairwise_distance(
-        adata=adata, meta=meta, bootstrap=bootstrap, thresh=thresh, **nei_kwargs
+        adata=adata,
+        meta=meta,
+        bootstrap=bootstrap,
+        noise_scale=noise_scale,
+        thresh=thresh,
+        **nei_kwargs,
     )
     result = ripser(
         distance_matrix=sparse_pairwise_distance_matrix.tocoo(copy=False),
@@ -90,3 +104,41 @@ def compute_boundary_matrix_data(
             triangles, meta.preprocess.num_vertices
         )
     return result, edge_ids, trig_ids, sparse_pairwise_distance_matrix, vertex_indices
+
+
+def compute_loop_homological_equivalence(
+    boundary_matrix_d1: "BoundaryMatrixD1",
+    loop_mask_a: np.ndarray,
+    loop_mask_b: np.ndarray,
+    n_pairs_check: int = 3,
+) -> tuple[list, list]:
+    """
+    Parameters
+    ---------
+    loop_mask_a: np.ndarray
+        Boolean mask of shape (n_a, n_edges); True where edge (row) is in the loop
+    loop_mask_b: np.ndarray
+        Boolean mask of shape (n_b, n_edges)
+    """
+    assert loop_mask_a.shape[1] == boundary_matrix_d1.shape[0]
+    assert loop_mask_b.shape[1] == boundary_matrix_d1.shape[0]
+
+    # in F2, sum is just xor
+    loop_sums = loop_mask_a[:, None, :] ^ loop_mask_b[None, :, :]
+    loop_sums = loop_sums.reshape(-1, loop_sums.shape[-1])
+    if loop_sums.shape[0] == 0:
+        return [], []
+    n_pairs_check = min(n_pairs_check, loop_sums.shape[0])
+
+    one_idx_b_list = [
+        np.flatnonzero(loop_sums[i]).astype(int).tolist() for i in range(n_pairs_check)
+    ]
+    results, solutions = solve_multiple_gf2(
+        one_ridx_A=boundary_matrix_d1.data[0],
+        one_cidx_A=boundary_matrix_d1.data[1],
+        nrow_A=boundary_matrix_d1.shape[0],
+        ncol_A=boundary_matrix_d1.shape[1],
+        one_idx_b_list=one_idx_b_list,
+    )
+
+    return results, solutions

src/scloop/data/__init__.py

@@ -1,2 +1,2 @@
 from .containers import HomologyData
-from .ripser_lib import RipserResults, get_boundary_matrix, ripser
+from .ripser_lib import RipserResults, get_boundary_matrix, ripser  # type: ignore[import-not-found]

src/scloop/data/analysis_containers.py

@@ -2,19 +2,16 @@
 from __future__ import annotations
 
 import numpy as np
-import pandas as pd
 from pydantic import ConfigDict, Field
 from pydantic.dataclasses import dataclass
 
 
-@dataclass(config=ConfigDict(arbitrary_types_allowed=True))
+@dataclass
 class BootstrapAnalysis:
-    loops_eidx_boot: list[list[list[np.ndarray]]] = Field(default_factory=list)
-    persistence_diagram_boot: list[np.ndarray] = Field(default_factory=list)
-    matching_df: list[pd.DataFrame] = Field(default_factory=list)
-    n_booted: int = 0
-    loop_rank: pd.DataFrame = Field(default_factory=pd.DataFrame)
-    parameters: dict = Field(default_factory=dict)
+    num_bootstraps: int = 0
+    persistence_diagrams: list[list] = Field(default_factory=list)
+    cocycles: list[list] = Field(default_factory=list)
+    loop_representatives: list[list[list[list[int]]]] = Field(default_factory=list)
 
 
 @dataclass(config=ConfigDict(arbitrary_types_allowed=True))