Mercurial > repos > goeckslab > squidpy_scatter
changeset 0:6fe0d4f464f4 draft
planemo upload for repository https://github.com/goeckslab/tools-mti/tree/main/tools/squidpy commit 721eaced787aa3b04d96ad91f6b4540f26b23949
| author | goeckslab |
|---|---|
| date | Thu, 11 Jul 2024 22:22:53 +0000 |
| parents | |
| children | b84c324b58bd |
| files | main_macros.xml squidpy_scatter.py squidpy_scatter.xml squidpy_spatial.py test-data/imc.h5ad test-data/imc_centrality_scores.png test-data/imc_co_occurrence.png test-data/imc_interaction_matrix.png test-data/imc_kmeans.h5ad test-data/imc_nhood_enrichment.png test-data/imc_ripley.png test-data/imc_sn.h5ad test-data/scatter_image.png |
| diffstat | 13 files changed, 328 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/main_macros.xml Thu Jul 11 22:22:53 2024 +0000 @@ -0,0 +1,64 @@ +<macros> + <token name="@TOOL_VERSION@">1.5.0</token> + <token name="@PROFILE@">20.01</token> + <token name="@VERSION_SUFFIX@">1</token> + + <xml name="macro_stdio"> + <stdio> + <exit_code range="1:" level="fatal" description="Error occurred. Please check Tool Standard Error" /> + </stdio> + </xml> + + <xml name="squidpy_requirements"> + <requirements> + <requirement type="package" version="@TOOL_VERSION@">squidpy</requirement> + </requirements> + </xml> + + <xml name="citations"> + <citations> + <citation type="doi">10.1038/s41592-021-01358-2</citation> + </citations> + </xml> + + <xml name="squidpy_spatial_options"> + <section name="options" title="Advanced Graph Options" expanded="false"> + <yield/> + <!-- <param argument="copy" type="hidden" value="false" help="If True, return the result, otherwise save it to the adata object." /> --> + </section> + </xml> + + <xml name="squidpy_plotting_options"> + <section name="plotting_options" title="Plotting Options" expanded="false"> + <yield /> + <!-- <param argument="legend_kwargs" type="text" value="" optional="true" label="Keyword arguments for matplotlib.pyplot.legend()" /> --> + <param argument="figsize" type="text" value="" optional="true" label="Size of the figure in inches" help="Optional. e.g.: (12, 12)." /> + <param argument="dpi" type="integer" value="" optional="true" label="Dots per inch" help="Optional" /> + </section> + </xml> + + <xml name="squidpy_plotting_option_palette"> + <param argument="palette" type="text" value="" optional="true" label="Categorical colormap for the clusters" help="Comma delimited for multiple. If None, use anndata.AnnData.uns ['{cluster_key}_colors'], if available." /> + </xml> + + <xml name="squidpy_plotting_options_more"> + <expand macro="squidpy_plotting_options"> + <yield /> + <param argument="annotate" type="boolean" checked="false" label="Whether to annotate the cells of the heatmap?" /> + <param argument="method" type="select" label="The linkage method to be used for dendrogram/clustering" help="see scipy.cluster.hierarchy.linkage()."> + <option value="none" selected="true">None</option> + <option value="single">single</option> + <option value="complete">complete</option> + <option value="average">average</option> + <option value="weighted">weighted</option> + <option value="centroid">centroid</option> + <option value="median">median</option> + <option value="ward">ward</option> + </param> + <param argument="title" type="text" value="" optional="true" label="The title of the plot" help="Optional." /> + <param argument="cmap" type="text" value="viridis" label="Continuous colormap to use" help="Refer to `matplotlib.pyplot.cmap`." /> + <!-- <param argument="cbar_kwargs" type="text" value="" label="Keyword arguments for matplotlib.figure.Figure.colorbar()" /> --> + <expand macro="squidpy_plotting_option_palette" /> + </expand> + </xml> +</macros>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/squidpy_scatter.py Thu Jul 11 22:22:53 2024 +0000 @@ -0,0 +1,87 @@ +import argparse +import ast +import json +import warnings + +import squidpy as sq +from anndata import read_h5ad + + +def main(inputs, output_plot): + + """ + inputs : str + File path to galaxy tool JSON inputs config file + output_plot: str + File path to save the plotting image + """ + warnings.simplefilter('ignore') + + # read inputs JSON + with open(inputs, 'r') as param_handler: + params = json.load(param_handler) + + # collapse param dict hierarchy, parse inputs + plot_opts = params.pop('plot_opts') + legend_opts = params.pop('legend_opts') + aes_opts = params.pop('aesthetic_opts') + options = {**params, **plot_opts, **legend_opts, **aes_opts} + + # read input anndata file + adata_fh = options.pop('anndata') + adata = read_h5ad(adata_fh) + + # ensure spatial coords in anndata.obsm['spatial'] + # if not, populate with user provided X/Y coord column names + x, y = options.pop('x_coord'), options.pop('y_coord') + if 'spatial' not in adata.obsm: + try: + adata.obsm['spatial'] = adata.obs[[x, y]].values + except Exception as e: + print(e) + + # scan thru tool params, + # replace None values, and reformat specific parameters + for k, v in options.items(): + if not isinstance(v, str): + continue + + if v in ('', 'none'): + options[k] = None + continue + + if k == 'groups': + options[k] = [e.strip() for e in v.split(',')] + + elif k == 'crop_coord': + # split str on commas into tuple of coords + # then nest in list (expected by squidpy function) + options[k] = [tuple([int(e.strip()) for e in v.split(',')])] + + elif k == 'figsize': + options[k] = ast.literal_eval(v) + + # not exposing this parameter for now. Only useful to change for ST data + # and can otherwise just be problematic. + # Explicitly setting to None is necessary to avoid an error + options['shape'] = None + + # call squidpy spatial scatter function, unpack tool params + sq.pl.spatial_scatter( + adata=adata, + save='image.png', + **options + ) + + +if __name__ == '__main__': + + aparser = argparse.ArgumentParser() + aparser.add_argument( + "-i", "--inputs", dest="inputs", required=True) + aparser.add_argument( + "-p", "--output_plot", dest="output_plot", required=False) + + args = aparser.parse_args() + + main(args.inputs, args.output_plot)
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/squidpy_scatter.xml Thu Jul 11 22:22:53 2024 +0000 @@ -0,0 +1,72 @@ +<tool id="squidpy_scatter" name="Create spatial scatterplot" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@"> + <description>with Squidpy</description> + <macros> + <import>main_macros.xml</import> + </macros> + <edam_operations> + <edam_operation>operation_3443</edam_operation> + </edam_operations> + <expand macro="squidpy_requirements"/> + <expand macro="macro_stdio" /> + <version_command>echo "@TOOL_VERSION@"</version_command> + <command detect_errors="aggressive"> + <![CDATA[ + export TQDM_DISABLE=True && + python '$__tool_directory__/squidpy_scatter.py' + --inputs '$inputs' + --output_plot image.png + ]]> + </command> + <configfiles> + <inputs name="inputs" data_style="paths"/> + </configfiles> + <inputs> + <param name="anndata" type="data" format="h5ad" label="Select the input anndata" /> + <param name="x_coord" type="text" value="X_centroid" label="Key for X spatial coordinate in adata.obs" /> + <param name="y_coord" type="text" value="Y_centroid" label="Key for Y spatial coordinate in adata.obs" /> + <param argument="color" type="text" value="phenotype" label="Key for annotations in adata.obs or variables/genes" /> + <section name="plot_opts" title="Advanced Plot Options" expanded="false"> + <param argument="groups" type="text" optional="true" label="For discrete annotation in color, select which values to plot" /> + <param argument="use_raw" type="boolean" checked="false" label="Use adata.raw.X instead of adata.X" /> + <param argument="crop_coord" type="text" optional="true" label="Coordinates to crop scatterplot by" help="x1,y1,x2,y2" /> + </section> + <section name="legend_opts" title="Legend Options" expanded="false"> + <param argument="legend_loc" type="text" value="right margin" optional="true" label="Location of the legend" /> + <param argument="legend_fontsize" type="integer" value="8" optional="true" label="Font size for the legend" /> + <param argument="legend_fontweight" type="text" value="bold" optional="true" label="Font weight for the legend" /> + <param argument="legend_fontoutline" type="integer" value="" optional="true" label="Font outline for the legend" /> + <param argument="legend_na" type="boolean" checked="true" label="Show NA values in the legend" /> + </section> + <section name="aesthetic_opts" title="Figure Aesthetic Options" expanded="false"> + <param argument="size" type="integer" value="3" label="Size of the scatter point/shape" /> + <param argument="alpha" type="float" min="0.01" max="1.0" value="1.0" label="Alpha value for scatter point/shape" /> + <param argument="cmap" type="text" value="viridis" optional="true" label="Matplotlib colormap for continuous annotations" help="Refer to `matplotlib.pyplot.cmap`." /> + <param argument="colorbar" type="boolean" checked="true" label="Show the colorbar" /> + <param argument="scalebar_dx" type="float" min="0.01" max="1.0" value="" optional="true" label="Size of one pixel in units specified by scalebar_units" /> + <param argument="scalebar_units" type="text" value="" optional="true" label="Units for scalebar" /> + <param argument="title" type="text" optional="true" label="Figure title" /> + <param argument="figsize" type="text" value="" optional="true" label="Size of the figure in inches" help="width,height"/> + <param argument="dpi" type="integer" value="300" optional="true" label="Dots per inch" /> + </section> + </inputs> + <outputs> + <data from_work_dir="figures/image.png" format="png" name="output_plot" label="Squidpy spatial scatterplot on ${on_string}" /> + </outputs> + <tests> + <test> + <param name="anndata" value="imc_sn.h5ad" ftype="h5ad" /> + <param name="color" value="cell type" /> + <output name="output_plot" file="scatter_image.png" compare="sim_size" delta="2000" /> + </test> + </tests> + <help> + <![CDATA[ +This is a limited implementation of the squidpy.pl.spatial_scatter() function + +See Squidpy documentation for more information: + +https://squidpy.readthedocs.io/en/stable/index.html + ]]> + </help> + <expand macro="citations" /> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/squidpy_spatial.py Thu Jul 11 22:22:53 2024 +0000 @@ -0,0 +1,105 @@ +import argparse +import ast +import json +import warnings + +import pandas as pd +import squidpy as sq +from anndata import read_h5ad + + +def main(inputs, anndata, output, output_plot): + """ + Parameter + --------- + inputs : str + File path to galaxy tool parameter. + anndata : str + File path to anndata containing phenotyping info. + output : str + File path to output. + output_plot: str or None + File path to save the plotting image. + """ + warnings.simplefilter('ignore') + + with open(inputs, 'r') as param_handler: + params = json.load(param_handler) + + adata = read_h5ad(anndata) + + if 'spatial' not in adata.obsm: + try: + adata.obsm['spatial'] = adata.obs[['X_centroid', 'Y_centroid']].values + except Exception as e: + print(e) + + tool = params['analyses']['selected_tool'] + tool_func = getattr(sq.gr, tool) + + options = params['analyses']['options'] + + for k, v in options.items(): + if not isinstance(v, str): + continue + + if v in ('', 'none'): + options[k] = None + continue + + if k == 'genes': # for spatial_autocorr and sepal + options[k] = [e.strip() for e in v.split(',')] + elif k == 'radius': # for spatial_neighbors + options[k] = ast.literal_eval(v) + elif k == 'interactions': # for ligrec + options[k] = pd.read_csv(v, sep="\t") + elif k == 'max_neighs': + options[k] = int(v) # for sepal + + cluster_key = params['analyses'].get('cluster_key') + if cluster_key: + tool_func(adata, cluster_key, **options) + else: + tool_func(adata, **options) + + if output_plot: + plotting_options = params['analyses']['plotting_options'] + for k, v in plotting_options.items(): + if not isinstance(v, str): + continue + + if v in ('', 'none'): + plotting_options[k] = None + continue + + if k == 'figsize': + options[k] = ast.literal_eval(v) + elif k in ('palette', 'score', 'source_groups', 'target_groups'): + options[k] = [e.strip() for e in v.split(',')] + elif k == 'means_range': # ligrec + v = v.strip() + if v[0] == '(': + v = v[1:] + if v[-1] == ')': + v = v[:-1] + options[k] = tuple([float(e.strip()) for e in v.split(',', 1)]) + + plotting_func = getattr(sq.pl, tool) + if cluster_key: + plotting_func(adata, cluster_key, save=output_plot, **plotting_options) + else: # TODO Remove this, since all plottings need cluster key + plotting_func(adata, save=output_plot, **plotting_options) + + adata.write(output) + + +if __name__ == '__main__': + aparser = argparse.ArgumentParser() + aparser.add_argument("-i", "--inputs", dest="inputs", required=True) + aparser.add_argument("-e", "--output", dest="output", required=True) + aparser.add_argument("-a", "--anndata", dest="anndata", required=True) + aparser.add_argument("-p", "--output_plot", dest="output_plot", required=False) + + args = aparser.parse_args() + + main(args.inputs, args.anndata, args.output, args.output_plot)