SpatialQuery 0.0.1__py3-none-any.whl

SpatialQuery/utils.py

@@ -0,0 +1,130 @@
+from collections import Counter
+from itertools import combinations
+
+import numpy as np
+import pandas as pd
+import anndata as ad
+from sklearn.preprocessing import LabelEncoder
+import scanpy as sc
+
+
+def maximal_patterns(fp,
+                     key: str = 'itemsets',
+                     ):
+    """
+    Retrieve the maximal patterns in provided frequent patterns.
+
+    Parameter
+    ---------
+    fp:
+        DataFrame with frequent patterns
+    key:
+        Column name representing the patterns.
+
+    Return
+    ------
+    A dataframe with the maximal patterns.
+    """
+    itemsets = fp[key].apply(frozenset)
+
+    # Find all subsets of each itemset
+    subsets = set()
+    for itemset in itemsets:
+        for r in range(1, len(itemset)):
+            subsets.update(frozenset(s) for s in combinations(itemset, r))
+
+    # Identify maximal patterns (itemsets that are not subsets of any other)
+    maximal_patterns = [tuple(sorted(itemset)) for itemset in itemsets if itemset not in subsets]
+    # maximal_patterns_ = [list(p) for p in maximal_patterns]
+
+    # Filter the original DataFrame to keep only the maximal patterns
+    return fp[fp['itemsets'].isin(maximal_patterns)].reset_index(drop=True)
+
+
+# TODO: query efficiency can be improved as in motif_enrichment_dist of single FOV, filtering cells
+def retrieve_niche_pattern_freq(fp, sp, ct, max_dist):
+    """
+    Retrieve frequency of each cell type in frequent pattern (fp) around
+    central cell type (ct) from single FOV (sp).
+
+    Parameters:
+    fp: List[str]
+        list of cell types
+    sp: spatial_query object
+        spatial query object for single FOV
+    ct: str
+        central cell type
+    max_dist: float
+        radius of neighborhood
+
+    Return:
+        AnnData with central cell type and neighboring cells in fp
+    """
+    if ct not in sp.labels.unique():
+        print(f"{ct} does not exist in FOV.")
+        return pd.DataFrame(columns=fp)
+
+    cinds = [i for i, label in enumerate(sp.labels) if label == ct]  # id of center cell type
+
+    # ct_pos = self.spatial_pos[cinds]
+    idxs = sp.kd_tree.query_ball_point(sp.spatial_pos, r=max_dist, return_sorted=False, workers=-1)
+
+    label_encoder = LabelEncoder()
+    int_labels = label_encoder.fit_transform(np.array(sp.labels))
+    int_ct = label_encoder.transform(np.array(ct, dtype=object, ndmin=1))
+
+    num_cells = len(idxs)
+    num_types = len(label_encoder.classes_)
+    idxs_filter = [np.array(ids)[np.array(ids) != i] for i, ids in enumerate(idxs)]
+    flat_neighbors = np.concatenate(idxs_filter)
+    row_indices = np.repeat(np.arange(num_cells), [len(neigh) for neigh in idxs_filter])
+    neighbor_labels = int_labels[flat_neighbors]
+
+    neighbor_matrix = np.zeros((num_cells, num_types), dtype=int)
+    np.add.at(neighbor_matrix, (row_indices, neighbor_labels), 1)
+
+    mask = int_labels == int_ct
+
+    motif_exc = [m for m in fp if m not in sp.labels.unique()]
+    if len(motif_exc) != 0:
+        print(f"Found no {motif_exc} in {sp.label_key}. Ignoring them.")
+    motif = [m for m in fp if m not in motif_exc]
+
+    int_motifs = label_encoder.transform(np.array(fp))
+
+    inds = np.where(np.all(neighbor_matrix[mask][:, int_motifs] > 0, axis=1))[0]
+    cind_with_motif = [cinds[i] for i in inds]  # id of ct with fp nearby
+
+    freqs = []
+    for id in cind_with_motif:
+        ns = idxs_filter[id]
+        freq_all = Counter(sp.labels[ns])
+        freq_fp = {ct: count / len(ns) for ct, count in freq_all.items() if ct in fp}
+        freqs.append(freq_fp)
+
+    return pd.DataFrame(freqs)
+
+def plot_niche_pattern_freq(freqs):
+    """
+    Heatmap plot of frequency of patterns (cell type compositions) in niche.
+
+    Parameter
+    ---------
+    freqs: Output of retrieve_niche_pattern_freq method.
+    """
+
+    for i, freq in freqs.items():
+        freq['FOV'] = f"normal_{i}"
+        freqs[i] = freq
+
+    freq_fp_normal = pd.concat(freqs)
+    freq_fp_normal.set_index(keys='FOV', drop=True, inplace=True)
+    fp_data = ad.AnnData(X=freq_fp_normal)
+    var = pd.DataFrame({'neighbor': freq_fp_normal.columns.tolist()})
+    var.index = freq_fp_normal.columns.tolist()
+    obs = pd.DataFrame({'FOV': freq_fp_normal.index.tolist()})
+    obs.index = freq_fp_normal.index.tolist()
+    fp_data.var = var
+    fp_data.obs = obs
+
+    sc.pl.heatmap(fp_data, var_names=fp_data.var_names, groupby='FOV', cmap='vlag')

spatialquery-0.0.1.dist-info/METADATA

@@ -0,0 +1,56 @@
+Metadata-Version: 2.4
+Name: SpatialQuery
+Version: 0.0.1
+Home-page: https://github.com/ShaokunAn/Spatial-Query
+Author: Shaokun An
+Author-email: shan12@bwh.harvard.edu
+License: MIT
+Classifier: Development Status :: 3 - Alpha
+Classifier: Intended Audience :: Developers
+Classifier: Programming Language :: Python :: 3
+Requires-Python: >=3.8
+Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: setuptools>=68.0.0
+Requires-Dist: anndata>=0.8.0
+Requires-Dist: pandas>=2.0.3
+Requires-Dist: scipy
+Requires-Dist: matplotlib>=3.7.5
+Requires-Dist: mlxtend>=0.23.1
+Requires-Dist: seaborn>=0.13.2
+Requires-Dist: scikit-learn>=1.3.2
+Dynamic: author
+Dynamic: author-email
+Dynamic: classifier
+Dynamic: description
+Dynamic: description-content-type
+Dynamic: home-page
+Dynamic: license
+Dynamic: license-file
+Dynamic: requires-dist
+Dynamic: requires-python
+
+SpatialQuery is a package for fast query of Spatial Transcriptomics data. 
+
+### Analysis of ST data in SpatialQuery
+
+With annotated ST data as input, SpatialQuery first builds a k-D tree based on spatial location in each FOV for fast query of neighboring cell compositions. It is composed of methods for single-FOV and multiple-FOVs.
+In single-FOV, it contains methods:
+
+- identify frequent patterns across FOV
+- identify frequent patterns around cell type of interest
+- identify statistically significant patterns around cell type of interest
+
+For multiple-FOVs data, it contains methods:
+
+- identify frequent patterns around cell type of interest in specified datasets
+- identify statistically significant patterns around cell type of interest in specified datasets
+- identify differential patterns across datasets
+
+### Installation
+
+```
+pip install -i https://test.pypi.org/simple/ SpatialQuery
+```
+
+For more details, please refer to `examples/SpatialQuery-example.ipynb`.

spatialquery-0.0.1.dist-info/RECORD

@@ -0,0 +1,9 @@
+SpatialQuery/__init__.py,sha256=m9euWH8hvuDnUAx3t9Fwxv9frDUFrGbzUMGjqOJbMa4,191
+SpatialQuery/spatial_query.py,sha256=U0XMNDcEwYp18HTdTlCD7j9Zwvokmy4hjPVdYvTYQxE,57127
+SpatialQuery/spatial_query_multiple_fov.py,sha256=xTcVT_0V3faIUry_d5YnXEBcJdR17CoLSbqSjPZtvRg,49287
+SpatialQuery/utils.py,sha256=PvBgrp7gp6wHIdluc3V-VM-tUIyfrP7FpbvBjVP5ZM0,4350
+spatialquery-0.0.1.dist-info/licenses/LICENSE,sha256=FWlbKlEJSYI5oSGvTThBwoRePIhTbb0YlvOqmNLeK78,1067
+spatialquery-0.0.1.dist-info/METADATA,sha256=qKh8GAa5JIdMo1ZNyqQQ9QBshzBbk6IzqTzv1H-XSM4,1840
+spatialquery-0.0.1.dist-info/WHEEL,sha256=pxyMxgL8-pra_rKaQ4drOZAegBVuX-G_4nRHjjgWbmo,91
+spatialquery-0.0.1.dist-info/top_level.txt,sha256=kuI1SAjBSJ-Y0QMWXCzZAs06drSn9rM_KH_iT0jG5hw,13
+spatialquery-0.0.1.dist-info/RECORD,,

spatialquery-0.0.1.dist-info/WHEEL

@@ -0,0 +1,5 @@
+Wheel-Version: 1.0
+Generator: setuptools (79.0.0)
+Root-Is-Purelib: true
+Tag: py3-none-any
+

spatialquery-0.0.1.dist-info/licenses/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 Shaokun An
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

spatialquery-0.0.1.dist-info/top_level.txt

@@ -0,0 +1 @@
+SpatialQuery