do some updates.

docs/source/Tutorials(Multi-sample)/ST_Gears.ipynb

@@ -254,7 +254,7 @@
    "source": [
     "## Running ST-Gears\n",
     "\n",
-    "`start_i` and `end_i` specify computing scope of samples, here are mean to compute sample 8 to sample 11.\n",
+    "`start_i` and `end_i` specify computing scope of samples, here means to compute sample 8 to sample 11.\n",
     "\n",
     "For more details refer to [ms_data.tl.st_gears](../content/stereo.algorithm.st_gears.StGears.main.html)."
    ]

docs/source/Tutorials/SpaTrack.rst

@@ -1,5 +1,9 @@
 SpaTrack
 ====================
+
+Introducion
+--------------------
+
 Trajectory inference (TI) provides important insights in understanding cell development and biological process.
 However, the integration of transcriptomic profiles and spatial locations to organize spatiotemporal cell orders is currently remaining challenges. 
 Here we introduce spaTrack, which effectively constructs cell trajectories from an optimal-transport matrix at single cell resolution, 
@@ -10,6 +14,22 @@ as well as reconstruct cell dynamics across multiple tissue sections in a time s
 spaTrack models the fate of a cell as a function of expression profile along the time points driven by transcription factors, 
 which facilitates the identification of key molecular regulators that govern cellular trajectories.
 
+Highlighted features
+---------------------
+
+1. reconstructs fine local trajectories from ST data.
+2. integrates spatial transition matrix of multiple samples to generate complete trajectories.
+3. traces cell trajectory across multiple tissue sections via direct mapping without integrating data.
+4. captures the driven factors of differentiation.
+5. could be extensively applied on both ST data and scRNA-seq data.
+6. requires lower computing memory and loads than RNA-velocity methods, making it a fast and effective option for TI study.
+
+Turorials
+---------------------
+
+We provide the following five tutorials as reference cases to illustrate the application of spaTrack in inferring cell trajectories 
+on ST data of single slices with a single starting point and multiple starting points, as well as ST data of multiple time slices, and scRNA-seq data.
+
 .. nbgallery::
 
     Apply_spaTrack_on_spatial_data_of_axolotl_brain_regeneration_after_injury

docs/source/content/032_API_StPipeline.rst

@@ -16,6 +16,7 @@ which is compromised of basic preprocessing, embedding, clustering, and so on.
    :toctree: .
 
     core.StPipeline
+    core.StPipeline.set_layer
     core.StPipeline.cal_qc
     core.StPipeline.filter_cells
     core.StPipeline.filter_genes

docs/source/content/06_Release_notes.rst

@@ -3,6 +3,80 @@ Release Notes
 
 .. role:: small
 
+Version 1.5.0
+------------------
+1.5.0 : 2024-11-08
+~~~~~~~~~~~~~~~~~~~
+
+.. _SpaTrack: ../Tutorials/SpaTrack.html
+.. |SpaTrack| replace:: **SpaTrack**
+
+.. _Layer: stereo.core.StPipeline.set_layer.html
+.. |Layer| replace:: **Layer**
+
+.. _st.tl.cal_qc: stereo.core.StPipeline.cal_qc.html
+.. |st.tl.cal_qc| replace:: `st.tl.cal_qc`
+
+.. _st.tl.filter_cells: stereo.core.StPipeline.filter_cells.html
+.. |st.tl.filter_cells| replace:: `st.tl.filter_cells`
+
+.. _st.tl.filter_genes: stereo.core.StPipeline.filter_genes.html
+.. |st.tl.filter_genes| replace:: `st.tl.filter_genes`
+
+.. _st.tl.log1p: stereo.core.StPipeline.log1p.html
+.. |st.tl.log1p| replace:: `st.tl.log1p`
+
+.. _st.tl.normalize_total: stereo.core.StPipeline.normalize_total.html
+.. |st.tl.normalize_total| replace:: `st.tl.normalize_total`
+
+.. _st.tl.scale: stereo.core.StPipeline.scale.html
+.. |st.tl.scale| replace:: `st.tl.scale`
+
+.. _st.tl.quantile: stereo.core.StPipeline.quantile.html
+.. |st.tl.quantile| replace:: `st.tl.quantile`
+
+.. _st.tl.disksmooth_zscore: stereo.core.StPipeline.disksmooth_zscore.html
+.. |st.tl.disksmooth_zscore| replace:: `st.tl.disksmooth_zscore`
+
+.. _st.tl.sctransform: stereo.core.StPipeline.sctransform.html
+.. |st.tl.sctransform| replace:: `st.tl.sctransform`
+
+.. _st.tl.highly_variable_genes: stereo.core.StPipeline.highly_variable_genes.html
+.. |st.tl.highly_variable_genes| replace:: `st.tl.highly_variable_genes`
+
+.. _st.tl.pca: stereo.core.StPipeline.pca.html
+.. |st.tl.pca| replace:: `st.tl.pca`
+
+.. _st.tl.find_marker_genes: stereo.core.StPipeline.find_marker_genes.html
+.. |st.tl.find_marker_genes| replace:: `st.tl.find_marker_genes`
+
+.. _st.plt.spatial_scatter: stereo.plots.PlotCollection.spatial_scatter.html
+.. |st.plt.spatial_scatter| replace:: `st.plt.spatial_scatter`
+
+Features:
+
+1. Addition of new algorithm |SpaTrack|_ for trajectory inference.
+2. Addition of |Layer|_ for saving expression matrices at different analysis stages, the functions that can use expression matrices in **Layer** as following:
+            * |st.tl.cal_qc|_
+            * |st.tl.filter_cells|_
+            * |st.tl.filter_genes|_
+            * |st.tl.log1p|_
+            * |st.tl.normalize_total|_
+            * |st.tl.scale|_
+            * |st.tl.quantile|_
+            * |st.tl.disksmooth_zscore|_
+            * |st.tl.sctransform|_
+            * |st.tl.highly_variable_genes|_
+            * |st.tl.pca|_
+            * |st.tl.find_marker_genes|_
+3. Merger of multiple samples can merge some analysis result in every single samples when data type is **StereoExpData**.
+4. |st.plt.spatial_scatter|_ supports setting base image as background to display simultaneously on the plot.
+
+BUG Fixes:
+
+1. Fixed the problem that the proportion of chondriogenes was calculated incorrectly when input data contains **geneID**.
+2. Fixed the problem that saving **MSData** into **h5mu** was failed after running `st.tl.highly_variable_genes`.
+
 Version 1.4.0
 ------------------
 1.4.0 : 2024-09-05

docs/source/index.rst

@@ -70,6 +70,80 @@ Workflow
 Latest Additions
 ------------------
 
+Version 1.5.0
+~~~~~~~~~~~~~~~~~~~
+1.5.0 : 2024-11-08
+
+.. _SpaTrack: Tutorials/SpaTrack.html
+.. |SpaTrack| replace:: **SpaTrack**
+
+.. _Layer: content/stereo.core.StPipeline.set_layer.html
+.. |Layer| replace:: **Layer**
+
+.. _st.tl.cal_qc: content/stereo.core.StPipeline.cal_qc.html
+.. |st.tl.cal_qc| replace:: `st.tl.cal_qc`
+
+.. _st.tl.filter_cells: content/stereo.core.StPipeline.filter_cells.html
+.. |st.tl.filter_cells| replace:: `st.tl.filter_cells`
+
+.. _st.tl.filter_genes: content/stereo.core.StPipeline.filter_genes.html
+.. |st.tl.filter_genes| replace:: `st.tl.filter_genes`
+
+.. _st.tl.log1p: content/stereo.core.StPipeline.log1p.html
+.. |st.tl.log1p| replace:: `st.tl.log1p`
+
+.. _st.tl.normalize_total: content/stereo.core.StPipeline.normalize_total.html
+.. |st.tl.normalize_total| replace:: `st.tl.normalize_total`
+
+.. _st.tl.scale: content/stereo.core.StPipeline.scale.html
+.. |st.tl.scale| replace:: `st.tl.scale`
+
+.. _st.tl.quantile: content/stereo.core.StPipeline.quantile.html
+.. |st.tl.quantile| replace:: `st.tl.quantile`
+
+.. _st.tl.disksmooth_zscore: content/stereo.core.StPipeline.disksmooth_zscore.html
+.. |st.tl.disksmooth_zscore| replace:: `st.tl.disksmooth_zscore`
+
+.. _st.tl.sctransform: content/stereo.core.StPipeline.sctransform.html
+.. |st.tl.sctransform| replace:: `st.tl.sctransform`
+
+.. _st.tl.highly_variable_genes: content/stereo.core.StPipeline.highly_variable_genes.html
+.. |st.tl.highly_variable_genes| replace:: `st.tl.highly_variable_genes`
+
+.. _st.tl.pca: content/stereo.core.StPipeline.pca.html
+.. |st.tl.pca| replace:: `st.tl.pca`
+
+.. _st.tl.find_marker_genes: content/stereo.core.StPipeline.find_marker_genes.html
+.. |st.tl.find_marker_genes| replace:: `st.tl.find_marker_genes`
+
+.. _st.plt.spatial_scatter: content/stereo.plots.PlotCollection.spatial_scatter.html
+.. |st.plt.spatial_scatter| replace:: `st.plt.spatial_scatter`
+
+Features:
+
+1. Addition of new algorithm |SpaTrack|_ for trajectory inference.
+2. Addition of |Layer|_ for saving expression matrices at different analysis stages, the functions that can use expression matrices in **Layer** as following:
+            * |st.tl.cal_qc|_
+            * |st.tl.filter_cells|_
+            * |st.tl.filter_genes|_
+            * |st.tl.log1p|_
+            * |st.tl.normalize_total|_
+            * |st.tl.scale|_
+            * |st.tl.quantile|_
+            * |st.tl.disksmooth_zscore|_
+            * |st.tl.sctransform|_
+            * |st.tl.highly_variable_genes|_
+            * |st.tl.pca|_
+            * |st.tl.find_marker_genes|_
+3. Merger of multiple samples can merge some analysis result in every single samples when data type is **StereoExpData**.
+4. |st.plt.spatial_scatter|_ supports setting base image as background to display simultaneously on the plot.
+
+BUG Fixes:
+
+1. Fixed the problem that the proportion of chondriogenes was calculated incorrectly when input data contains **geneID**.
+2. Fixed the problem that saving **MSData** into **h5mu** was failed after running `st.tl.highly_variable_genes`.
+
+
 Version 1.4.0
 ~~~~~~~~~~~~~~~~~~~
 1.4.0 : 2024-09-05
@@ -116,25 +190,6 @@ BUG Fixes:
 2. Fixed the problem that `st.io.read_gef` is incompatible with those **GEF** files that contain **gene names** ending with **'_{number}'** (like **'ABC_123'**).
 3. Upgraded **gefpy** to latest for fixing the error that **gene names** are lost after running **CellCorrection**.
 
-Version 1.3.0
-~~~~~~~~~~~~~~~~~~~
-1.3.0 : 2024-05-31
-
-Features:
-
-1. Addition of `MSData.tl.st_gears <Tutorials(Multi-sample)/ST_Gears.html>`_ for spatial alignment of **Multi-sample**.
-2. `High Resolution Matrix Export <Tutorials/High_Resolution_Export.html>`_ can support both **GEF** and **GEM** files.
-3. Addition of parameters `min_count` and `max_count` for `st.tl.filter_genes <content/stereo.core.StPipeline.filter_genes.html>`_.
-4. `MSData.integrate <content/stereo.core.ms_data.MSData.integrate.html>`_ can be compatible with sparse matrix when `MSData.var_type` is `union`.
-5. Addition of `MSData.tl.set_scope_and_mode <content/stereo.core.ms_pipeline.MSDataPipeLine.set_scope_and_mode.html>`_ to set `scope` and `mode` globally on **Multi-sample** analysis.
-6. Addition of `MSData.plt.ms_spatial_scatter <content/stereo.plots.PlotMsSpatialScatter.ms_spatial_scatter.html>`_ to plot spatial scatter plot for each **sample** in **Multi-sample** separately.
-
-BUG Fixes:
-
-1. Fixed the problem that `st.io.read_gem` is incompatible with **GEM** files containing **geneID**.
-2. Fixed the bug of losing part of metadata when writing **StereoExpData** / **MSData** into **Stereo-h5ad** or **h5ms** file.
-3. Fixed the incompatibility problem with **AnnData** when performing `st.tl.sctransform`.
-
 
 .. toctree::
     :titlesonly:

stereo/algorithm/get_niche.py

@@ -42,9 +42,8 @@ def main(
         assert cluster_1 != cluster_2, "cluster_1 can not equal to cluster_2."
 
         data_full = self.stereo_exp_data
-        cluster = self.pipeline_res[cluster_res_key]
-        data_1, _ = filter_by_clusters(data_full, cluster_res=cluster, groups=cluster_1, inplace=False)
-        data_2, _ = filter_by_clusters(data_full, cluster_res=cluster, groups=cluster_2, inplace=False)
+        data_1 = filter_by_clusters(data_full, cluster_res_key=cluster_res_key, groups=cluster_1, inplace=False)
+        data_2 = filter_by_clusters(data_full, cluster_res_key=cluster_res_key, groups=cluster_2, inplace=False)
 
         coord_1 = data_1.position.astype(np.int64)
         coord_2 = data_2.position.astype(np.int64)
@@ -76,7 +75,7 @@ def main(
             neighbors = np.zeros((n1, n12), dtype=int)
             shift = np.zeros(n1, dtype=np.float64)
 
-            cluster_label = cluster['group']
+            cluster_label = self.pipeline_res[cluster_res_key]['group']
             info_entropy = np.zeros(n1, dtype=np.float64)
 
             for i in range(n1):

stereo/core/ms_data.py

@@ -276,6 +276,8 @@ class _MSDataStruct(object):
     def __check_data_list(self, data_list):
         if not isinstance(data_list, list):
             raise TypeError('data_list must be a list object')
+        if len(data_list) == 0:
+            return data_list
         first_data = data_list[0]
         for data in data_list[1:]:
             if type(data) != type(first_data):

stereo/core/result.py

@@ -693,7 +693,7 @@ def _get_marker_genes_res(self, name):
             'control_groups': marker_genes_result['params']['reference'],
             'corr_method': marker_genes_result['params']['corr_method'],
             'use_raw': marker_genes_result['params']['use_raw'],
-            'layer': marker_genes_result['params']['layer']
+            'layer': marker_genes_result['params']['layer'] if 'layer' in marker_genes_result['params'] else None,
         }
         if 'marker_genes_res_key' in marker_genes_result['params']:
             marker_genes_result_reconstructed['parameters']['marker_genes_res_key'] = \

stereo/plots/plot_collection.py

@@ -866,7 +866,7 @@ def umap(
                 height=height,
                 **kwargs)
         else:
-            self.data.array2sparse()
+            # self.data.array2sparse()
             if gene_names is None:
                 raise ValueError('gene name must be set if cluster_key is None')
             if isinstance(gene_names, str):

stereo/plots/scatter.py

@@ -215,23 +215,24 @@ def base_scatter(
 
     """  # noqa
     if not ax:
-        # if width is None or height is None:
-        #     figsize = (7, 7)
-        # else:
-        #     width = width / 100 if width >= 100 else 7
-        #     height = height / 100 if height >= 100 else 7
-        #     figsize = (width, height)
+        min_x, max_x = np.min(x), np.max(x)
+        min_y, max_y = np.min(y), np.max(y)
+        data_width = (max_x - min_x) / 100
+        data_height = (max_y - min_y) / 100
+        default_width = 7
+        default_height = 7 * data_height / data_width
         if width is None:
-            width = 7
+            width = default_width
         else:
-            width = width / 100 if width >= 100 else 7
+            width = width / 100 if width >= 100 else default_width
         if height is None:
-            height = 7
+            height = default_height
         else:
-            height = height / 100 if height >= 100 else 7
+            height = height / 100 if height >= 100 else default_height
         _, ax = plt.subplots(figsize=(width, height))
     dot_size = PLOT_SCATTER_SIZE_FACTOR / len(hue) if dot_size is None else dot_size
     # add a color bar
+
 
     if invert_y:
         ax.invert_yaxis()
@@ -276,27 +277,17 @@ def base_scatter(
         sm = plt.cm.ScalarMappable(cmap=cmap, norm=norm)
         sm.set_array([])
         ax.figure.colorbar(sm)
-        # if ax.legend_ is not None:
-        #     ax.legend_.remove()
     else:
         from natsort import natsorted
         import collections
         g = natsorted(set(hue))
         if hue_order is None:
             hue_order = g
-        # if isinstance(palette, (dict, collections.OrderedDict)):
-        #     palette = [palette[i] for i in g if i in palette]
-        # if len(palette) < len(g):
-        #     colors = stereo_conf.get_colors(palette, n=len(g))
-        # else:
-        #     colors = palette
-        # color_dict = collections.OrderedDict(dict([(g[i], colors[i]) for i in range(len(g))]))
         colors = stereo_conf.get_colors(palette, n=len(g), order=hue_order)
         color_dict = dict(zip(hue_order, colors))
         sns.scatterplot(x=x, y=y, hue=hue, hue_order=hue_order, linewidth=0, marker=marker,
                         palette=color_dict, size=hue, sizes=(dot_size, dot_size), ax=ax, alpha=foreground_alpha)
         handles, labels = ax.get_legend_handles_labels()
-        # ax.legend_.remove()
         legd = ax.legend(handles, labels, ncol=legend_ncol, bbox_to_anchor=(1.02, 1),
                   loc='upper left', borderaxespad=0, frameon=False)
         for lh in legd.legendHandles:
@@ -372,14 +363,21 @@ def multi_scatter(
     hue_length = len(hue) if isinstance(hue, list) else hue.shape[0]
     ncols = min(ncols, hue_length)
     nrows = np.ceil(hue_length / ncols).astype(int)
-    # each panel will have the size of rcParams['figure.figsize']
-    if width is None or height is None:
-        figsize = (ncols * 10, nrows * 8)
+    min_x, max_x = np.min(x), np.max(x)
+    min_y, max_y = np.min(y), np.max(y)
+    data_width = (max_x - min_x) / 100
+    data_height = (max_y - min_y) / 100
+    default_width = 8 + 2
+    default_height = 8 * data_height / data_width
+    if width is None:
+        width = ncols * default_width
+    else:
+        width = width / 100 if width >= 100 else ncols * default_width
+    if height is None:
+        height = nrows * default_height
     else:
-        width = width / 100 if width >= 100 else ncols * 8
-        height = height / 100 if height >= 100 else nrows * 8
-        figsize = (width, height)
-    fig = plt.figure(figsize=figsize)
+        height = height / 100 if height >= 100 else nrows * default_height
+    fig = plt.figure(figsize=(width, height))
     left = 0.2 / ncols
     bottom = 0.13 / nrows
     axs = gridspec.GridSpec(
@@ -412,6 +410,10 @@ def multi_scatter(
                      data_resolution=data_resolution,
                      **kwargs
                      )
+    # if width is not None:
+    #     fig.set_figwidth(width)
+    # if height is not None:
+    #     fig.set_figheight(height)
     return fig