pgalign 1.1.0__tar.gz → 2.0.0__tar.gz

{pgalign-1.1.0 → pgalign-2.0.0}/CITATION.cff

@@ -2,7 +2,7 @@ cff-version: 1.2.0
 message: "If you use PGAlign in your research, please cite the accompanying paper."
 title: "PGAlign: peak group alignment of mass spectrometry data"
 type: software
-version: "1.1.0"
+version: "2.0.0"
 authors:
   - family-names: Deng
     given-names: Anji

{pgalign-1.1.0/pgalign.egg-info → pgalign-2.0.0}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: pgalign
-Version: 1.1.0
-Summary: PGAlign: peak group alignment of mass spectrometry data for imaging and spatial metabolomics.
+Version: 2.0.0
+Summary: Peak group alignment of mass spectrometry imaging data onto a reference m/z grid.
 Author: Anji Deng, Yuyang Zhang
 Author-email: Qihuang Zhang <qihuang.zhang@mcgill.ca>
 Maintainer-email: Qihuang Zhang <qihuang.zhang@mcgill.ca>

{pgalign-1.1.0 → pgalign-2.0.0}/pgalign/PeakCalling.py

@@ -136,8 +136,13 @@ class PeakCalling(object):
         self.jointcluster = [clusteri for clusteri in self.jointcluster if clusteri[-1] < mzboundary]
 
     def importpeak(self, peakUnk, peakRef):
+        """Use peaks called elsewhere in place of Step 1.
+
+        :param peakUnk: indices of the peak points in the unknown spectrum.
+        :param peakRef: indices of the peak points in the reference spectrum.
+        """
         self.peakUnk = np.zeros(self.mz_valueUnk.shape[0], dtype=bool)
-        self.peakRef = self.peakUnk
+        self.peakRef = np.zeros(self.mz_valueRef.shape[0], dtype=bool)
         for i in range(peakUnk.shape[0]):
             self.peakUnk[peakUnk[i]] = True
         for j in range(peakRef.shape[0]):
@@ -187,7 +192,12 @@ class PeakCalling_single(PeakCalling):
         self.jointcluster = [clusteri for clusteri in self.jointcluster if clusteri[-1] < mzboundary]
 
     def importpeak(self, peakUnk, peakRef):
-        self.peakRef = self.peakUnk
+        """Use peaks called elsewhere in place of Step 1; ``peakUnk`` is ignored.
+
+        :param peakUnk: unused, kept for the signature of ``PeakCalling.importpeak``.
+        :param peakRef: indices of the peak points in the reference spectrum.
+        """
+        self.peakRef = np.zeros(self.mz_valueRef.shape[0], dtype=bool)
         for j in range(peakRef.shape[0]):
             self.peakRef[peakRef[j]] = True
         self.meanspectrumRefpeak = self.meanspectrumRef[self.peakRef]

{pgalign-1.1.0 → pgalign-2.0.0}/pgalign/__init__.py

@@ -11,22 +11,12 @@ The four pipeline steps in the paper map to these public objects:
 align() runs all four with the manuscript's defaults and returns the aligned dataset.
 """
 
-__version__ = "1.1.0"
+__version__ = "2.0.0"
 
 from .util import (
-    prefilter_cells,
-    prefilter_genes,
-    prefilter_specialgenes,
+    gridding,
     save_clusterresults,
-    rotate,
-    group_range,
-    in_range_lookup,
-    comp_clusters,
-    find_nearest,
-    get_unk_comp_clusters,
     spectrum_save,
-    GKernal,
-    gridding,
 )
 from .AnnDataMALDI import (
     AnnDataMALDI,
@@ -71,17 +61,7 @@ __all__ = [
     # Toy data
     "datasets",
     # Utilities
-    "prefilter_cells",
-    "prefilter_genes",
-    "prefilter_specialgenes",
+    "gridding",
     "save_clusterresults",
-    "rotate",
-    "group_range",
-    "in_range_lookup",
-    "comp_clusters",
-    "find_nearest",
-    "get_unk_comp_clusters",
     "spectrum_save",
-    "GKernal",
-    "gridding",
 ]

{pgalign-1.1.0 → pgalign-2.0.0}/pgalign/util.py

@@ -1,44 +1,10 @@
-"""Utility helpers used by PGAlign's peak-calling and alignment routines."""
+"""Utilities. ``gridding``, ``save_clusterresults`` and ``spectrum_save`` are public;
+the other helpers serve the step classes and are not exported from ``pgalign``."""
 
-import logging
 import math
 
 import numpy as np
 import pandas as pd
-import scanpy as sc
-
-logger = logging.getLogger(__name__)
-
-
-def prefilter_cells(adata, min_counts=None, max_counts=None, min_genes=200, max_genes=None):
-    if min_genes is None and min_counts is None and max_genes is None and max_counts is None:
-        raise ValueError('Provide one of min_counts, min_genes, max_counts or max_genes.')
-    id_tmp = np.asarray([True] * adata.shape[0], dtype=bool)
-    id_tmp = np.logical_and(id_tmp, sc.pp.filter_cells(adata.X, min_genes=min_genes)[0]) if min_genes is not None else id_tmp
-    id_tmp = np.logical_and(id_tmp, sc.pp.filter_cells(adata.X, max_genes=max_genes)[0]) if max_genes is not None else id_tmp
-    id_tmp = np.logical_and(id_tmp, sc.pp.filter_cells(adata.X, min_counts=min_counts)[0]) if min_counts is not None else id_tmp
-    id_tmp = np.logical_and(id_tmp, sc.pp.filter_cells(adata.X, max_counts=max_counts)[0]) if max_counts is not None else id_tmp
-    adata._inplace_subset_obs(id_tmp)
-    adata.raw = sc.pp.log1p(adata, copy=True)
-    logger.debug("adata.raw.var_names.is_unique = %s", adata.raw.var_names.is_unique)
-
-
-def prefilter_genes(adata, min_counts=None, max_counts=None, min_cells=10, max_cells=None):
-    if min_cells is None and min_counts is None and max_cells is None and max_counts is None:
-        raise ValueError('Provide one of min_counts, min_genes, max_counts or max_genes.')
-    id_tmp = np.asarray([True] * adata.shape[1], dtype=bool)
-    id_tmp = np.logical_and(id_tmp, sc.pp.filter_genes(adata.X, min_cells=min_cells)[0]) if min_cells is not None else id_tmp
-    id_tmp = np.logical_and(id_tmp, sc.pp.filter_genes(adata.X, max_cells=max_cells)[0]) if max_cells is not None else id_tmp
-    id_tmp = np.logical_and(id_tmp, sc.pp.filter_genes(adata.X, min_counts=min_counts)[0]) if min_counts is not None else id_tmp
-    id_tmp = np.logical_and(id_tmp, sc.pp.filter_genes(adata.X, max_counts=max_counts)[0]) if max_counts is not None else id_tmp
-    adata._inplace_subset_var(id_tmp)
-
-
-def prefilter_specialgenes(adata, Gene1Pattern="ERCC", Gene2Pattern="MT-"):
-    id_tmp1 = np.asarray([not str(name).startswith(Gene1Pattern) for name in adata.var_names], dtype=bool)
-    id_tmp2 = np.asarray([not str(name).startswith(Gene2Pattern) for name in adata.var_names], dtype=bool)
-    id_tmp = np.logical_and(id_tmp1, id_tmp2)
-    adata._inplace_subset_var(id_tmp)
 
 
 def save_clusterresults(cluster, filepath):
@@ -110,8 +76,8 @@ def spectrum_save(mz, intensity, name):
     originaltable.to_csv("{name}.csv".format(name=name))
 
 
-def GKernal(x0, x, y, sig=1.):
-    """Gaussian kernel smoother evaluated at ``x0``."""
+def nadaraya_watson(x0, x, y, sig=1.):
+    """Nadaraya-Watson estimate of ``y`` at ``x0``, with a Gaussian kernel of bandwidth ``sig``."""
     ax = x - x0
     Kernal = np.exp(-0.5 * np.square(ax) / np.square(sig))
     Weighted = Kernal * y
@@ -127,7 +93,7 @@ def gridding(mz, intensity, start=None, end=None, increment=0.125, epsilon=0.025
     target_grid = np.arange(start, end, increment)
     intensity_out = np.zeros(target_grid.shape[0])
     for i in range(target_grid.shape[0]):
-        calculation = GKernal(target_grid[i], mz, intensity, sig=epsilon)
+        calculation = nadaraya_watson(target_grid[i], mz, intensity, sig=epsilon)
         if calculation == calculation:
             intensity_out[i] = calculation
         else:

{pgalign-1.1.0 → pgalign-2.0.0/pgalign.egg-info}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: pgalign
-Version: 1.1.0
-Summary: PGAlign: peak group alignment of mass spectrometry data for imaging and spatial metabolomics.
+Version: 2.0.0
+Summary: Peak group alignment of mass spectrometry imaging data onto a reference m/z grid.
 Author: Anji Deng, Yuyang Zhang
 Author-email: Qihuang Zhang <qihuang.zhang@mcgill.ca>
 Maintainer-email: Qihuang Zhang <qihuang.zhang@mcgill.ca>

{pgalign-1.1.0 → pgalign-2.0.0}/pgalign.egg-info/SOURCES.txt

@@ -25,6 +25,7 @@ tests/test_ambiguity.py
 tests/test_checks.py
 tests/test_cli.py
 tests/test_deprecation.py
+tests/test_docs.py
 tests/test_import.py
 tests/test_peak_calling.py
 tests/test_readers.py

{pgalign-1.1.0 → pgalign-2.0.0}/pyproject.toml

@@ -5,7 +5,7 @@ build-backend = "setuptools.build_meta"
 [project]
 name = "pgalign"
 dynamic = ["version"]
-description = "PGAlign: peak group alignment of mass spectrometry data for imaging and spatial metabolomics."
+description = "Peak group alignment of mass spectrometry imaging data onto a reference m/z grid."
 readme = "README.md"
 license = "MIT"
 license-files = ["LICENSE"]

pgalign-2.0.0/tests/test_docs.py

@@ -0,0 +1,52 @@
+"""Run the Python examples of README.md and docs/ as written, on the bundled toy data."""
+
+import re
+from pathlib import Path
+
+import numpy as np
+import pytest
+
+ROOT = Path(__file__).resolve().parents[1]
+PAGES = ["README.md"] + sorted(str(p.relative_to(ROOT)) for p in (ROOT / "docs").glob("*.md"))
+# Examples that read the reader's own files cannot run here.
+NEEDS_FILES = ('read_imzml("', 'read_scils_csv("', 'read_h5ad("')
+# The README step-by-step example leaves the inputs to the reader; supply the toy data.
+PLACEHOLDER = ("MALDIdataAnn1 = ...   # unknown spectrum\n", "MALDIdataAnn2 = ...   # reference spectrum\n")
+TOY = "MALDIdataAnn1, MALDIdataAnn2 = pg.datasets.load_mouse_pancreas_toy()\n"
+
+
+def _blocks(page):
+    text = (ROOT / page).read_text(encoding="utf-8")
+    return [b for b in re.findall(r"```python\n(.*?)```", text, flags=re.S) if not any(s in b for s in NEEDS_FILES)]
+
+
+def _run(page):
+    namespace = {}
+    for block in _blocks(page):
+        exec(block.replace(PLACEHOLDER[0], TOY).replace(PLACEHOLDER[1], ""), namespace)
+        if "res" in namespace and "res.summary()" in block:
+            assert namespace["res"].summary() in block, f"{page}: the summary in a comment no longer matches"
+    return namespace
+
+
+@pytest.mark.parametrize("page", [p for p in PAGES if (ROOT / p).exists() and _blocks(p)])
+def test_examples_run_as_written(page):
+    _run(page)
+
+
+def test_readme_examples_align_onto_the_reference_grid():
+    if not (ROOT / "README.md").exists():
+        pytest.skip("README.md is not in this checkout")
+    ns = _run("README.md")
+    aligned, ref = ns["MALDIdataAnn1_aligned"], ns["MALDIdataAnn2"]
+    np.testing.assert_array_equal(aligned.var["m/z"].to_numpy(), ref.var["m/z"].to_numpy()[:aligned.n_vars])
+    assert len(ns["table"]) > 0
+
+
+def test_tutorial_step_by_step_equals_align():
+    if not (ROOT / "docs" / "tutorial.md").exists():
+        pytest.skip("docs/ is not in this checkout")
+    ns = _run("docs/tutorial.md")
+    # the toy shift is a rigid +2, so the matrix alone cannot tell the two peak callers apart
+    np.testing.assert_array_equal(ns["align"].align_group, ns["res"].steps.align_group)
+    np.testing.assert_array_equal(np.asarray(ns["aligned"].X), np.asarray(ns["res"].aligned.X))

{pgalign-1.1.0 → pgalign-2.0.0}/tests/test_import.py

@@ -19,11 +19,25 @@ def test_public_api_present():
         "MALDI_SIM",
         "PGmzalign",
         "gridding",
-        "GKernal",
+        "save_clusterresults",
+        "spectrum_save",
+        "read_imzml",
+        "read_scils_csv",
+        "read_h5ad",
+        "write_h5ad",
     ]:
         assert hasattr(pg, name), f"pgalign is missing public name: {name}"
 
 
+def test_removed_names_are_gone():
+    """Single-cell leftovers were deleted in 2.0; internal helpers stay in pgalign.util only."""
+    for name in ["prefilter_cells", "prefilter_genes", "prefilter_specialgenes", "GKernal",
+                 "rotate", "group_range", "in_range_lookup", "comp_clusters", "find_nearest",
+                 "get_unk_comp_clusters"]:
+        assert not hasattr(pg, name), f"pgalign still exports {name}"
+    assert callable(pg.util.in_range_lookup) and callable(pg.util.nadaraya_watson)
+
+
 def test_paper_aligned_methods_exist():
     """Method names from the current PGAlign manuscript are exposed on the right classes."""
     assert hasattr(pg.PeakCalling, "peak_calling")

pgalign-2.0.0/tests/test_peak_calling.py

@@ -0,0 +1,43 @@
+"""Smoke test for PGAlign Steps 1 and 2 on synthetic data."""
+
+import numpy as np
+
+import pgalign as pg
+
+
+def test_peak_calling_and_grouping_produce_clusters(tiny_anndata):
+    unk, ref = tiny_anndata
+    pc = pg.PeakCalling(unk, ref)
+    pc.peak_calling(threshold=0.9)
+    pc.peak_grouping(percentile=0.9)
+    # At least one joint peak group must be detected on the four-peak fixture.
+    assert len(pc.jointcluster) > 0
+    # Each joint peak group is a list of m/z values.
+    for group in pc.jointcluster:
+        assert len(group) >= 1
+
+
+def test_importpeak_keeps_the_two_spectra_apart(tiny_anndata):
+    unk, ref = tiny_anndata
+    pc = pg.PeakCalling(unk, ref)
+    pc.importpeak(np.array([10, 20]), np.array([30, 40]))
+    assert np.flatnonzero(pc.peakUnk).tolist() == [10, 20]
+    assert np.flatnonzero(pc.peakRef).tolist() == [30, 40]
+
+
+def test_imported_peaks_give_the_same_groups_as_step_1(tiny_anndata):
+    unk, ref = tiny_anndata
+    called = pg.PeakCalling(unk, ref)
+    called.peak_calling(threshold=0.9)
+    called.peak_grouping(percentile=0.9)
+    imported = pg.PeakCalling(unk, ref)
+    imported.importpeak(np.flatnonzero(called.peakUnk), np.flatnonzero(called.peakRef))
+    imported.peak_grouping(percentile=0.9)
+    assert imported.jointcluster == called.jointcluster
+
+
+def test_single_spectrum_importpeak(tiny_anndata):
+    _, ref = tiny_anndata
+    single = pg.PeakCalling_single(ref)
+    single.importpeak(None, np.array([30, 40]))
+    assert np.flatnonzero(single.peakRef).tolist() == [30, 40]

pgalign-1.1.0/tests/test_peak_calling.py

@@ -1,15 +0,0 @@
-"""Smoke test for PGAlign Steps 1 and 2 on synthetic data."""
-
-import pgalign as pg
-
-
-def test_peak_calling_and_grouping_produce_clusters(tiny_anndata):
-    unk, ref = tiny_anndata
-    pc = pg.PeakCalling(unk, ref)
-    pc.peak_calling(threshold=0.9)
-    pc.peak_grouping(percentile=0.9)
-    # At least one joint peak group must be detected on the four-peak fixture.
-    assert len(pc.jointcluster) > 0
-    # Each joint peak group is a list of m/z values.
-    for group in pc.jointcluster:
-        assert len(group) >= 1