Skip to content

Add Leiden clustering and update documentation #334

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
newmana merged 7 commits into from
Sep 17, 2025
Merged

Add Leiden clustering and update documentation #334

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
2089d25
Add leiden clustering.
newmana Sep 12, 2025
e4d48b4
Fix documentation and slight refactor.
newmana Sep 17, 2025
5b2c545
Fix style.
newmana Sep 17, 2025
5ab4bb8
Update metadata.
newmana Sep 17, 2025
7447f7c
Update history.
newmana Sep 17, 2025
3c1c5bb
Add types, fix warnings.
newmana Sep 17, 2025
502a69f
Update version.
newmana Sep 17, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions HISTORY.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,11 @@
History
=======

1.1.2 (2025-09-17)
------------------
* Add Leiden clustering wrapper.
* Fix documentation, refactor code in spatial.SME.

1.1.1 (2025-07-07)
------------------
* Support Python 3.10.x
Expand Down
6 changes: 6 additions & 0 deletions docs/release_notes/1.1.2.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,6 @@
1.1.2 `2025-09-17`
~~~~~~~~~~~~~~~~~~~~~~~~~

.. rubric:: Features
* Add Leiden clustering wrapper.
* Fix documentation, refactor code in spatial.SME.
2 changes: 2 additions & 0 deletions docs/release_notes/index.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,8 @@
Release Notes
===================================================

.. include:: 1.1.2.rst

.. include:: 1.1.1.rst

.. include:: 0.4.6.rst
Expand Down
4 changes: 2 additions & 2 deletions pyproject.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"

[project]
name = "stlearn"
version = "1.1.1"
version = "1.1.2"
authors = [
{name = "Genomics and Machine Learning lab", email = "andrew.newman@uq.edu.au"},
]
Expand All @@ -14,7 +14,7 @@ license = {text = "BSD license"}
requires-python = "~=3.10.0"
keywords = ["stlearn"]
classifiers = [
"Development Status :: 2 - Pre-Alpha",
"Development Status :: 5 - Production/Stable",
"Intended Audience :: Developers",
"License :: OSI Approved :: BSD License",
"Natural Language :: English",
Expand Down
3 changes: 2 additions & 1 deletion requirements.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,4 +11,5 @@ tensorflow==2.14.1
keras==2.14.0
types-tensorflow>=2.8.0
imageio==2.37.0
scipy==1.11.4
scipy==1.11.4
scikit-learn==1.7.0
5 changes: 3 additions & 2 deletions stlearn/spatial/SME/__init__.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,6 @@
from .impute import SME_impute0, pseudo_spot
from .normalize import SME_normalize
from .pseudo_spot import pseudo_spot
from .sme_impute0 import SME_impute0
from .sme_normalize import SME_normalize

__all__ = [
"SME_normalize",
Expand Down
128 changes: 60 additions & 68 deletions stlearn/spatial/SME/_weighting_matrix.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,7 +1,9 @@
import math
from typing import Literal

import numpy as np
from anndata import AnnData
from sklearn.linear_model import LinearRegression # type: ignore
from sklearn.metrics import pairwise_distances
from tqdm import tqdm

Expand All @@ -17,16 +19,9 @@
]


def calculate_weight_matrix(
adata: AnnData,
adata_imputed: AnnData | None = None,
pseudo_spots: bool = False,
platform: _PLATFORM = "Visium",
) -> AnnData | None:
import math

from sklearn.linear_model import LinearRegression

def row_col_by_platform(
adata, platform
) -> tuple[LinearRegression, LinearRegression, float]:
rate: float
if platform == "Visium":
img_row = adata.obs["imagerow"]
Expand All @@ -46,64 +41,61 @@ def calculate_weight_matrix(
{platform!r} does not support.
"""
)

reg_row = LinearRegression().fit(array_row.values.reshape(-1, 1), img_row)

reg_col = LinearRegression().fit(array_col.values.reshape(-1, 1), img_col)

if pseudo_spots and adata_imputed:
pd = pairwise_distances(
adata_imputed.obs[["imagecol", "imagerow"]],
adata.obs[["imagecol", "imagerow"]],
metric="euclidean",
)
unit = math.sqrt(reg_row.coef_**2 + reg_col.coef_**2)
pd_norm = np.where(pd >= unit, 0, 1)

md = 1 - pairwise_distances(
adata_imputed.obsm["X_morphology"],
adata.obsm["X_morphology"],
metric="cosine",
)
md[md < 0] = 0

adata_imputed.uns["physical_distance"] = pd_norm
adata_imputed.uns["morphological_distance"] = md

adata_imputed.uns["weights_matrix_all"] = (
adata_imputed.uns["physical_distance"]
* adata_imputed.uns["morphological_distance"]
)

else:
pd = pairwise_distances(adata.obs[["imagecol", "imagerow"]], metric="euclidean")
unit = math.sqrt(reg_row.coef_**2 + reg_col.coef_**2)
pd_norm = np.where(pd >= rate * unit, 0, 1)

md = 1 - pairwise_distances(adata.obsm["X_morphology"], metric="cosine")
md[md < 0] = 0

gd = 1 - pairwise_distances(adata.obsm["X_pca"], metric="correlation")
adata.uns["gene_expression_correlation"] = gd
adata.uns["physical_distance"] = pd_norm
adata.uns["morphological_distance"] = md

adata.uns["weights_matrix_all"] = (
adata.uns["physical_distance"]
* adata.uns["morphological_distance"]
* adata.uns["gene_expression_correlation"]
)
adata.uns["weights_matrix_pd_gd"] = (
adata.uns["physical_distance"] * adata.uns["gene_expression_correlation"]
)
adata.uns["weights_matrix_pd_md"] = (
adata.uns["physical_distance"] * adata.uns["morphological_distance"]
)
adata.uns["weights_matrix_gd_md"] = (
adata.uns["gene_expression_correlation"]
* adata.uns["morphological_distance"]
)
return adata
regression = LinearRegression()
reg_row: LinearRegression = regression.fit(array_row.values.reshape(-1, 1), img_row) # type: ignore
reg_col: LinearRegression = regression.fit(array_col.values.reshape(-1, 1), img_col) # type: ignore
return reg_col, reg_row, rate


def weight_matrix(adata, platform):
reg_col, reg_row, rate = row_col_by_platform(adata, platform)
pd = pairwise_distances(adata.obs[["imagecol", "imagerow"]], metric="euclidean")
unit = math.sqrt(reg_row.coef_[0] ** 2 + reg_col.coef_[0] ** 2)
pd_norm = np.where(pd >= rate * unit, 0, 1)
md = 1 - pairwise_distances(adata.obsm["X_morphology"], metric="cosine")
md[md < 0] = 0
gd = 1 - pairwise_distances(adata.obsm["X_pca"], metric="correlation")
adata.uns["gene_expression_correlation"] = gd
adata.uns["physical_distance"] = pd_norm
adata.uns["morphological_distance"] = md
adata.uns["weights_matrix_all"] = (
adata.uns["physical_distance"]
* adata.uns["morphological_distance"]
* adata.uns["gene_expression_correlation"]
)
adata.uns["weights_matrix_pd_gd"] = (
adata.uns["physical_distance"] * adata.uns["gene_expression_correlation"]
)
adata.uns["weights_matrix_pd_md"] = (
adata.uns["physical_distance"] * adata.uns["morphological_distance"]
)
adata.uns["weights_matrix_gd_md"] = (
adata.uns["gene_expression_correlation"] * adata.uns["morphological_distance"]
)


def weight_matrix_imputed(adata, adata_imputed, platform):
reg_col, reg_row, _ = row_col_by_platform(adata, platform)

pd = pairwise_distances(
adata_imputed.obs[["imagecol", "imagerow"]],
adata.obs[["imagecol", "imagerow"]],
metric="euclidean",
)
unit = math.sqrt(reg_row.coef_[0] ** 2 + reg_col.coef_[0] ** 2)
pd_norm = np.where(pd >= unit, 0, 1)
md = 1 - pairwise_distances(
adata_imputed.obsm["X_morphology"],
adata.obsm["X_morphology"],
metric="cosine",
)
md[md < 0] = 0
adata_imputed.uns["physical_distance"] = pd_norm
adata_imputed.uns["morphological_distance"] = md
adata_imputed.uns["weights_matrix_all"] = (
adata_imputed.uns["physical_distance"]
* adata_imputed.uns["morphological_distance"]
)


def impute_neighbour(
Expand Down
80 changes: 4 additions & 76 deletions stlearn/spatial/SME/impute.py → stlearn/spatial/SME/pseudo_spot.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,79 +12,10 @@
from ._weighting_matrix import (
_PLATFORM,
_WEIGHTING_MATRIX,
calculate_weight_matrix,
impute_neighbour,
weight_matrix_imputed,
)


def SME_impute0(
adata: AnnData,
use_data: str = "raw",
weights: _WEIGHTING_MATRIX = "weights_matrix_all",
platform: _PLATFORM = "Visium",
copy: bool = False,
) -> AnnData | None:
"""\
using spatial location (S), tissue morphological feature (M) and gene
expression (E) information to impute missing values

Parameters
----------
adata
Annotated data matrix.
use_data
input data, can be `raw` counts or log transformed data
weights
weighting matrix for imputation.
if `weights_matrix_all`, matrix combined all information from spatial
location (S), tissue morphological feature (M) and gene expression (E)
if `weights_matrix_pd_md`, matrix combined information from spatial
location (S), tissue morphological feature (M)
platform
`Visium` or `Old_ST`
copy
Return a copy instead of writing to adata.
Returns
-------
Anndata
"""
adata = adata.copy() if copy else adata

if use_data == "raw":
if isinstance(adata.X, csr_matrix):
count_embed = adata.X.toarray()
elif isinstance(adata.X, np.ndarray):
count_embed = adata.X
elif isinstance(adata.X, pd.Dataframe):
count_embed = adata.X.values
else:
raise ValueError(
f"""\
{type(adata.X)} is not a valid type.
"""
)
else:
count_embed = adata.obsm[use_data]

calculate_weight_matrix(adata, platform=platform)

impute_neighbour(adata, count_embed=count_embed, weights=weights)

imputed_data = adata.obsm["imputed_data"].astype(float)
mask = count_embed != 0
count_embed_ = count_embed.astype(float)
count_embed_[count_embed_ == 0] = np.nan
adjusted_count_matrix = np.nanmean(np.array([count_embed_, imputed_data]), axis=0)
adjusted_count_matrix[mask] = count_embed[mask]

key_added = use_data + "_SME_imputed"
adata.obsm[key_added] = adjusted_count_matrix

print("The data adjusted by SME is added to adata.obsm['" + key_added + "']")

return adata if copy else None


_COPY = Literal["pseudo_spot_adata", "combined_adata"]


Expand All @@ -98,9 +29,8 @@ def pseudo_spot(
copy: _COPY = "pseudo_spot_adata",
) -> AnnData | None:
"""\
using spatial location (S), tissue morphological feature (M) and gene
expression (E) information to impute gap between spots and increase resolution
for gene detection
Improve spatial resolution by imputing (creating) new spots from existing ones
using spatial, morphological, and expression (SME) information.

Parameters
----------
Expand Down Expand Up @@ -306,9 +236,7 @@ def pseudo_spot(
else:
count_embed = adata.obsm[use_data]

calculate_weight_matrix(
adata, pseudo_spot_adata, pseudo_spots=True, platform=platform
)
weight_matrix_imputed(adata, pseudo_spot_adata, platform=platform)

impute_neighbour(pseudo_spot_adata, count_embed=count_embed, weights=weights)

Expand Down
Loading