Mercurial > repos > iuc > spapros_evaluation

comparison evaluation.xml @ 0:0c1b0a3b19df draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload for repository https://github.com/galaxyproject/tools-iuc/tree/master/tools/spapros/ commit aed7fe13fa0ed09d77a31eeecaf3ec3fba7eed3b
author iuc
date Mon, 16 Sep 2024 11:37:34 +0000
parents
children
comparison
equal deleted inserted replaced
-1:000000000000 0:0c1b0a3b19df
1 <tool id="spapros_evaluation" name="Evaluation" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@profile@">
2 <description>of marker genes with spapros</description>
3 <macros>
4 <import>macros.xml</import>
5 </macros>
6 <expand macro="requirements">
7 </expand>
8 <expand macro="version_command"/>
9 <command detect_errors="exit_code"><![CDATA[
10 @CMD@
11 ]]></command>
12 <configfiles>
13 <configfile name="script_file"><![CDATA[
14 @CMD_imports@
15 @CMD_read_inputs@
16
17 mpl.rcParams['figure.dpi'] = $figure_options.dpi
18 plt.rcParams["font.size"] = $figure_options.fontsize
19
20 header_probeset='infer'
21 #if $header_probeset == 'not_included':
22 header_probeset=None
23 #end if
24
25 ## Probe set generation
26 probeset = pd.read_csv('$probeset', sep='\t', index_col=0, header=header_probeset)
27 probeset = [x for x in probeset.values.flatten().tolist() if not pd.isna(x)]
28
29 ## Reference set generation
30 methods_pre="#echo ','.join(["%s" % $s for $i, $s in enumerate($methods_reference)])#"
31 methods=methods_pre.split(',')
32
33 reference_sets = sp.se.select_reference_probesets(
34 adata,
35 n=$nreference,
36 #if $genes_key != '':
37 genes_key='$genes_key',
38 #else:
39 genes_key=None,
40 #end if
41 #if $obs_key != '':
42 obs_key='$obs_key',
43 #end if
44 #if $seeds != '':
45 seeds=[$seeds],
46 #end if
47 methods=methods,
48 verbosity=0
49 )
50
51 ## Parameter setting
52
53 custom_params = {
54 'data': {
55 'name': '$probeset_name',
56 'celltype_key': '$celltype_key'
57 },
58 'metrics': {
59 #if $method.method == 'plot_summary' or $method.method == 'plot_cluster_similarity':
60 'cluster_similarity': {
61 'ns': [$method.ns_start, $method.ns_end],
62 'AUC_borders': [#echo ",".join(["[%s, %s]" % ($s.auc_borders_start, $s.auc_borders_end) for $i, $s in enumerate($method.series_auc_borders)])#]
63 },
64 #end if
65 #if $method.method == 'plot_summary' or $method.method == 'plot_knn_overlap':
66 'knn_overlap': {
67 'ks': [#echo ",".join([ "%s" % $s.k for $i, $s in enumerate($method.knn) ])#]
68 },
69 #end if
70 #if $method.method == 'plot_confusion_matrix':
71 'forest_clfs': {
72 'ct_key': '$celltype_key',
73 #if $method.method == 'plot_summary':
74 'threshold': $method.forest_clfs_threshold
75 #end if
76 },
77 #end if
78 #if $method.method == 'plot_summary':
79 #if $method.select_marker_corr.use_marker_corr == 'True':
80 'marker_corr': {
81 'per_celltype': $method.select_marker_corr.per_celltype,
82 'per_marker': $method.select_marker_corr.per_marker,
83 #if $method.select_marker_corr.select_per_celltype_min_mean.use_per_celltype_min_mean == 'True'
84 'per_celltype_min_mean': $method.select_marker_corr.select_per_celltype_min_mean.per_celltype_min_mean,
85 #end if
86 #if $method.select_marker_corr.select_per_marker_min_mean.use_per_marker_min_mean == 'True'
87 'per_marker_min_mean': $method.select_marker_corr.select_per_marker_min_mean.per_marker_min_mean
88 #end if
89 },
90 #end if
91 #end if
92 #if $method.method == 'plot_marker_corr':
93 'marker_corr': {
94 'per_celltype': $method.per_celltype,
95 'per_marker': $method.per_marker,
96 #if $method.select_per_celltype_min_mean.use_per_celltype_min_mean == 'True'
97 'per_celltype_min_mean': $method.select_per_celltype_min_mean.per_celltype_min_mean,
98 #else:
99 'per_celltype_min_mean': None,
100 #end if
101 #if $method.select_per_marker_min_mean.use_per_marker_min_mean == 'True'
102 'per_marker_min_mean': $method.select_per_marker_min_mean.per_marker_min_mean
103 #else:
104 'per_marker_min_mean': None
105 #end if
106 },
107 #end if
108 #if $method.method == 'plot_summary':
109 'gene_corr': {
110 'threshold': $method.gene_corr_threshold
111 }
112 #end if
113 }
114 }
115
116 ## Setup evaluator
117
118 #if $method.method == 'plot_summary' or $method.method == 'plot_marker_corr':
119
120 header_markerset='infer'
121 feature_marker=dict()
122
123 #if $method.method == 'plot_summary':
124 #if $method.select_marker_corr.header_markerset == 'not_included' :
125 header_markerset=None
126 #end if
127 feature_marker = {key: [v for v in list(value.values()) if pd.notna(v)] for key, value in pd.read_csv('$method.select_marker_corr.markerset', sep='\t', index_col=0, header=header_markerset).to_dict(orient='index').items()}
128 #end if
129 #if $method.method == 'plot_marker_corr':
130 #if $method.header_markerset == 'not_included':
131 header_markerset=None
132 #end if
133 feature_marker = {key: [v for v in list(value.values()) if pd.notna(v)] for key, value in pd.read_csv('$method.markerset', sep='\t', index_col=0, header=header_markerset).to_dict(orient='index').items()}
134 #end if
135
136 #end if
137
138 evaluator = sp.ev.ProbesetEvaluator(
139 adata,
140 metrics_params=custom_params,
141 scheme="full",
142 verbosity=0,
143 #if $method.method == 'plot_summary' or $method.method == 'plot_marker_corr':
144 marker_list=feature_marker,
145 #end if
146 results_dir=None
147 )
148
149 ## Evaluation reference set
150 for set_id, df in reference_sets.items():
151 gene_set = df[df["selection"]].index.to_list()
152 evaluator.evaluate_probeset(gene_set, set_id=set_id)
153
154 ## Evaluation probe set
155 evaluator.evaluate_probeset(probeset, set_id='$probeset_name')
156
157 ## Execution of method ##
158
159 #if $method.method == 'plot_summary':
160 evaluator.plot_summary(
161 @CMD_plot@
162 )
163 #end if
164
165 #if $method.method == 'plot_confusion_matrix':
166 evaluator.plot_confusion_matrix(
167 @CMD_plot@
168 )
169 #end if
170
171 #if $method.method == 'plot_coexpression':
172 evaluator.plot_coexpression(
173 @CMD_plot@
174 )
175 #end if
176
177 #if $method.method == 'plot_cluster_similarity':
178 evaluator.plot_cluster_similarity(
179 @CMD_plot@
180 )
181 #end if
182
183 #if $method.method == 'plot_knn_overlap':
184 evaluator.plot_knn_overlap(
185 @CMD_plot@
186 )
187 #end if
188
189 ## plot_marker_corr does not allow for the show keyword
190 #if $method.method == 'plot_marker_corr':
191 evaluator.plot_marker_corr(
192 #if $method.set_ids != 'all' and $method.set_ids != '':
193 set_ids=[$method.set_ids],
194 #end if
195 save='plot.$format'
196 )
197 #end if
198
199 ]]></configfile>
200 </configfiles>
201 <inputs>
202 <expand macro="inputs_anndata"/>
203 <expand macro="param_plot_format"/>
204 <param name="probeset" type="data" format="tabular" label="Probeset tabular file with rows=conditions (e.g., celltypes) and column=features (e.g., genes)"/>
205 <param name="header_probeset" type="select" optional="false" label="Header in the list of probes?">
206 <option value="included">Header incldued</option>
207 <option value="not_included">Header not included</option>
208 </param>
209 <param argument="nreference" type="integer" value="10" min="1" optional="false" label="Number of selected genes to generate a reference set"/>
210 <param name="methods_reference" type="select" multiple="True" optional="false" label="Select the methods to genereate a reference probeset">
211 <option value="random" selected="true">Random</option>
212 <option value="PCA" selected="false">PCA</option>
213 <option value="DE" selected="false">Differential expressed features based on wilxocon rank genes test (DE)</option>
214 <option value="HVG" selected="false">Highly variable features (HVG)</option>
215 </param>
216 <param name="genes_key" type="text" optional="true" label="adata.var key for subset of preselected genes to run the selections" help="This is typically highly_variable_genes. Leave empty to not subset genes."/>
217 <param name="obs_key" type="text" optional="true" label="Column name of adata.obs for which marker scores are calculated" help="Only required for method DE."/>
218 <param name="seeds" type="text" optional="true" label="List of random integer seeds (comma seperated, e.g., 123,999,22)" help="For each seed, one random gene set is selected if random in methods. Leave it empty to for one random set with a random seed."/>
219 <param name="probeset_name" type="text" value="probeset" optional="false" label="Name for your probeset that should be displayed in plots"/>
220 <param name="celltype_key" type="text" value="celltype" optional="false" label="Column name of adata.obs that represents groups of your cells (e.g., celltype)"/>
221 <conditional name="method">
222 <param argument="method" type="select" optional="false" label="Method of spapros">
223 <option value="plot_summary">Plot statistic summary, using 'evaluator.plot_summary'</option>
224 <option value="plot_confusion_matrix">Plot confusion matrix, using 'evaluator.plot_confusion_matrix'</option>
225 <option value="plot_coexpression">Plot coexpression heatmap, using 'evaluator.plot_coexpression'</option>
226 <option value="plot_cluster_similarity">Plot cluster similariy, using 'evaluator.plot_cluster_similarity'</option>
227 <option value="plot_knn_overlap">Plot knn overlap, using 'evaluator.plot_knn_overlap'</option>
228 <option value="plot_marker_corr">Plot featue marker correlation, using 'evaluator.plot_marker_corr'</option>
229 </param>
230 <when value="plot_summary">
231 <expand macro="set_ids"/>
232 <param name="forest_clfs_threshold" type="float" value="0.8" min="0.0" max="1.0" label="Assesses how many cell types (%) can be predicted with an accuracy of at least threshold" help="The metric gives an idea about how many cell types can be identified with high confidence with the given gene set."/>
233 <param name="gene_corr_threshold" type="float" value="0.8" min="0.0" max="1.0" label="Percentage of features (e.g., genes) that have a maximum correlation of less than threshold with all other features" help="This metric gives an idea about how many features show unique expression profiles in the gene set."/>
234 <param name="ns_start" type="integer" value="5" min="1" label="The minimum number of leiden clusters clusters" help="Clusterings are calculated with different leiden resolutions to genertate clusterings of n = ns_start to ns_end clusters."/>
235 <param name="ns_end" type="integer" value="21" min="1" label="The maximum number of leiden clusters clusters" help="Clusterings are calculated with different leiden resolutions to genertate clusterings of n = ns_start to ns_end clusters."/>
236 <repeat name="series_auc_borders" title="Calculates nmi AUCs over given borders">
237 <param name="auc_borders_start" type="integer" value="15" min="1" label="Calculates nmi over n ranges auc_borders_start to auc_borders_end" help="Defined border shouldn't exceed values in nmis."/>
238 <param name="auc_borders_end" type="integer" value="20" min="1" label="Calculates nmi over n ranges auc_borders_start to auc_borders_end" help="Defined border shouldn't exceed values in nmis."/>
239 </repeat>
240 <repeat name="knn" title="Calculate knn graphs for each k">
241 <param name="k" type="integer" value="10" min="1" label="Includes nearest neighbors for all k"/>
242 </repeat>
243 <conditional name="select_marker_corr">
244 <param name="use_marker_corr" type="select" label="Do you want to calculate the correlation between your probeset and marker features?">
245 <option value="False">No</option>
246 <option value="True">Yes</option>
247 </param>
248 <when value="True">
249 <expand macro="param_markerset"/>
250 <param name="per_celltype" type="boolean" truevalue="True" falsevalue="False" label="Wether to return columns with per cell type max correlations" checked="true"/>
251 <param name="per_marker" type="boolean" truevalue="True" falsevalue="False" label="Wether to return columns with per marker max correlations" checked="true"/>
252 <conditional name="select_per_celltype_min_mean">
253 <param name="use_per_celltype_min_mean" type="select" label="Add a column for correlation per cell type that only takes into accounts markers with mean expression > per_celltype_min_mean">
254 <option value="False">No</option>
255 <option value="True">Yes</option>
256 </param>
257 <when value="True">
258 <param name="per_celltype_min_mean" type="float" value="0.0" min="0.0" label="Set per_celltype_min_mean"/>
259 </when>
260 <when value="False"/>
261 </conditional>
262 <conditional name="select_per_marker_min_mean">
263 <param name="use_per_marker_min_mean" type="select" label="Add a column for correlation per cell type that only takes into accounts markers with mean expression > per_celltype_min_mean">
264 <option value="False">No</option>
265 <option value="True">Yes</option>
266 </param>
267 <when value="True">
268 <param name="per_marker_min_mean" type="float" value="0.0" min="0.0" label="Set per_marker_min_mean"/>
269 </when>
270 <when value="False"/>
271 </conditional>
272 </when>
273 <when value="False"/>
274 </conditional>
275 </when>
276 <when value="plot_confusion_matrix">
277 <expand macro="set_ids"/>
278 </when>
279 <when value="plot_coexpression">
280 <expand macro="set_ids"/>
281 </when>
282 <when value="plot_cluster_similarity">
283 <expand macro="set_ids"/>
284 <param name="ns_start" type="integer" value="5" min="1" label="The minimum number of leiden clusters clusters" help="Clusterings are calculated with different leiden resolutions to genertate clusterings of n = ns_start to ns_end clusters."/>
285 <param name="ns_end" type="integer" value="21" min="1" label="The maximum number of leiden clusters clusters" help="Clusterings are calculated with different leiden resolutions to genertate clusterings of n = ns_start to ns_end clusters."/>
286 <repeat name="series_auc_borders" title="Calculates nmi AUCs over given borders">
287 <param name="auc_borders_start" type="integer" value="15" min="1" label="Calculates nmi over n ranges auc_borders_start to auc_borders_end" help="Defined border shouldn't exceed values in nmis."/>
288 <param name="auc_borders_end" type="integer" value="20" min="1" label="Calculates nmi over n ranges auc_borders_start to auc_borders_end" help="Defined border shouldn't exceed values in nmis."/>
289 </repeat>
290 </when>
291 <when value="plot_knn_overlap">
292 <expand macro="set_ids"/>
293 <repeat name="knn" title="Calculate knn graphs for each k">
294 <param name="k" type="integer" value="10" min="1" label="Includes nearest neighbors for all k"/>
295 </repeat>
296 </when>
297 <when value="plot_marker_corr">
298 <expand macro="param_markerset"/>
299 <expand macro="set_ids"/>
300 <param name="per_celltype" type="boolean" truevalue="True" falsevalue="False" label="Wether to return columns with per cell type max correlations" checked="true"/>
301 <param name="per_marker" type="boolean" truevalue="True" falsevalue="False" label="Wether to return columns with per marker max correlations" checked="true"/>
302 <conditional name="select_per_celltype_min_mean">
303 <param name="use_per_celltype_min_mean" type="select" label="Add a column for correlation per cell type that only takes into accounts markers with mean expression > per_celltype_min_mean">
304 <option value="False">No</option>
305 <option value="True">Yes</option>
306 </param>
307 <when value="True">
308 <param name="per_celltype_min_mean" type="float" value="0.0" min="0.0" label="Set per_celltype_min_mean"/>
309 </when>
310 <when value="False"/>
311 </conditional>
312 <conditional name="select_per_marker_min_mean">
313 <param name="use_per_marker_min_mean" type="select" label="Add a column for correlation per cell type that only takes into accounts markers with mean expression > per_celltype_min_mean">
314 <option value="False">No</option>
315 <option value="True">Yes</option>
316 </param>
317 <when value="True">
318 <param name="per_marker_min_mean" type="float" value="0.0" min="0.0" label="Set per_marker_min_mean"/>
319 </when>
320 <when value="False"/>
321 </conditional>
322 </when>
323 </conditional>
324 <section name="figure_options" title="Figure Output Options" expanded="false">
325 <param argument="dpi" type="integer" value="300" min="1" label="Dpi of figures"/>
326 <param argument="fontsize" type="integer" value="100" min="1" label="Font size of figures"/>
327 </section>
328 <expand macro="inputs_common_advanced"/>
329 </inputs>
330 <outputs>
331 <data name="out_png" format="png" from_work_dir="*.png" label="PNG plot from ${tool.name} (${method.method}) on ${on_string}">
332 <filter>format == 'png'</filter>
333 </data>
334 <data name="out_pdf" format="pdf" from_work_dir="*.pdf" label="PDF plot from ${tool.name} (${method.method}) on ${on_string}">
335 <filter>format == 'pdf'</filter>
336 </data>
337 <data name="out_svg" format="svg" from_work_dir="*.svg" label="SVG plot from ${tool.name} (${method.method}) on ${on_string}">
338 <filter>format == 'svg'</filter>
339 </data>
340 <expand macro="hidden_outputs"/>
341 </outputs>
342 <tests>
343 <test expect_num_outputs="2">
344 <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
345 <param name="format" value="png"/>
346 <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
347 <param name="header_probeset" value="included"/>
348 <param name="nreference" value="30"/>
349 <param name="methods_reference" value="random,PCA,DE,HVG"/>
350 <param name="genes_key" value="highly_variable"/>
351 <param name="obs_key" value="celltype"/>
352 <param name="seeds" value="123,124"/>
353 <conditional name="method">
354 <param name="method" value="plot_summary"/>
355 <param name="set_ids" value="all"/>
356 <param name="forest_clfs_threshold" value="0.8"/>
357 <param name="gene_corr_threshold" value="0.8"/>
358 <param name="ns_start" value="5"/>
359 <param name="ns_end" value="21"/>
360 <repeat name="series_auc_borders">
361 <param name="auc_borders_start" value="7"/>
362 <param name="auc_borders_end" value="14"/>
363 </repeat>
364 <repeat name="series_auc_borders">
365 <param name="auc_borders_start" value="15"/>
366 <param name="auc_borders_end" value="20"/>
367 </repeat>
368 <repeat name="knn">
369 <param name="k" value="5"/>
370 </repeat>
371 <repeat name="knn">
372 <param name="k" value="10"/>
373 </repeat>
374 <conditional name="select_marker_corr">
375 <param name="use_marker_corr" value="True"/>
376 <param name="markerset" value="marker.tsv"/>
377 <param name="header_markerset" value="not_included"/>
378 <param name="per_celltype" value="True"/>
379 <param name="per_marker" value="True"/>
380 <conditional name="select_per_celltype_min_mean">
381 <param name="use_per_celltype_min_mean" value="False"/>
382 </conditional>
383 <conditional name="select_per_marker_min_mean">
384 <param name="use_per_marker_min_mean" value="True"/>
385 <param name="per_marker_min_mean" value="0.025"/>
386 </conditional>
387 </conditional>
388 </conditional>
389 <param name="show_log" value="true" />
390 <output name="hidden_output">
391 <assert_contents>
392 <has_text_matching expression="'name': 'probeset',"/>
393 <has_text_matching expression="'per_celltype': True,"/>
394 <has_text_matching expression="'per_marker': True,"/>
395 <has_text_matching expression="'per_marker_min_mean': 0.025"/>
396 <has_text_matching expression="evaluator.plot_summary"/>
397 <has_text_matching expression="save='plot.png',"/>
398 </assert_contents>
399 </output>
400 <output name="out_png">
401 <assert_contents>
402 <has_image_width width="3253" delta="2"/>
403 <has_image_height height="1446" delta="2"/>
404 </assert_contents>
405 </output>
406 </test>
407 <test expect_num_outputs="2">
408 <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
409 <param name="format" value="png"/>
410 <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
411 <param name="header_probeset" value="included"/>
412 <param name="nreference" value="30"/>
413 <param name="methods_reference" value="random,PCA,DE,HVG"/>
414 <param name="genes_key" value="highly_variable"/>
415 <param name="obs_key" value="celltype"/>
416 <param name="seeds" value="123,124"/>
417 <conditional name="method">
418 <param name="method" value="plot_confusion_matrix"/>
419 <param name="set_ids" value="all"/>
420 </conditional>
421 <param name="show_log" value="true" />
422 <output name="hidden_output">
423 <assert_contents>
424 <has_text_matching expression="'name': 'probeset',"/>
425 <has_text_matching expression="'ct_key': 'celltype',"/>
426 <has_text_matching expression="evaluator.plot_confusion_matrix"/>
427 <has_text_matching expression="save='plot.png',"/>
428 </assert_contents>
429 </output>
430 <output name="out_png">
431 <assert_contents>
432 <has_image_width width="4560" delta="2"/>
433 <has_image_height height="1859" delta="2"/>
434 </assert_contents>
435 </output>
436 </test>
437 <test expect_num_outputs="2">
438 <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
439 <param name="format" value="png"/>
440 <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
441 <param name="header_probeset" value="included"/>
442 <param name="nreference" value="30"/>
443 <param name="methods_reference" value="random,PCA,DE,HVG"/>
444 <param name="genes_key" value="highly_variable"/>
445 <param name="obs_key" value="celltype"/>
446 <param name="seeds" value="123,124"/>
447 <conditional name="method">
448 <param name="method" value="plot_coexpression"/>
449 <param name="set_ids" value="all"/>
450 </conditional>
451 <param name="show_log" value="true" />
452 <output name="hidden_output">
453 <assert_contents>
454 <has_text_matching expression="'name': 'probeset',"/>
455 <has_text_matching expression="evaluator.plot_coexpression"/>
456 <has_text_matching expression="save='plot.png',"/>
457 </assert_contents>
458 </output>
459 <output name="out_png">
460 <assert_contents>
461 <has_image_width width="5412" delta="2"/>
462 <has_image_height height="3463" delta="2"/>
463 </assert_contents>
464 </output>
465 </test>
466 <test expect_num_outputs="2">
467 <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
468 <param name="format" value="png"/>
469 <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
470 <param name="header_probeset" value="included"/>
471 <param name="nreference" value="30"/>
472 <param name="methods_reference" value="random,PCA,DE,HVG"/>
473 <param name="genes_key" value="highly_variable"/>
474 <param name="obs_key" value="celltype"/>
475 <param name="seeds" value="123,124"/>
476 <conditional name="method">
477 <param name="method" value="plot_cluster_similarity"/>
478 <param name="set_ids" value="all"/>
479 <param name="ns_start" value="3"/>
480 <param name="ns_end" value="20"/>
481 <repeat name="series_auc_borders">
482 <param name="auc_borders_start" value="7"/>
483 <param name="auc_borders_end" value="14"/>
484 </repeat>
485 <repeat name="series_auc_borders">
486 <param name="auc_borders_start" value="15"/>
487 <param name="auc_borders_end" value="20"/>
488 </repeat>
489 </conditional>
490 <param name="show_log" value="true" />
491 <output name="hidden_output">
492 <assert_contents>
493 <has_text_matching expression="'name': 'probeset',"/>
494 <has_text_matching expression="evaluator.plot_cluster_similarity"/>
495 <has_text_matching expression="save='plot.png',"/>
496 </assert_contents>
497 </output>
498 <output name="out_png">
499 <assert_contents>
500 <has_image_width width="3223" delta="2"/>
501 <has_image_height height="1406" delta="2"/>
502 </assert_contents>
503 </output>
504 </test>
505 <test expect_num_outputs="2">
506 <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
507 <param name="format" value="png"/>
508 <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
509 <param name="header_probeset" value="included"/>
510 <param name="nreference" value="30"/>
511 <param name="methods_reference" value="random,PCA,DE,HVG"/>
512 <param name="genes_key" value="highly_variable"/>
513 <param name="obs_key" value="celltype"/>
514 <param name="seeds" value="123,124"/>
515 <conditional name="method">
516 <param name="method" value="plot_knn_overlap"/>
517 <param name="set_ids" value="all"/>
518 <repeat name="knn">
519 <param name="k" value="6"/>
520 </repeat>
521 <repeat name="knn">
522 <param name="k" value="11"/>
523 </repeat>
524 </conditional>
525 <param name="show_log" value="true" />
526 <output name="hidden_output">
527 <assert_contents>
528 <has_text_matching expression="'name': 'probeset',"/>
529 <has_text_matching expression="evaluator.plot_knn_overlap"/>
530 <has_text_matching expression="save='plot.png',"/>
531 </assert_contents>
532 </output>
533 <output name="out_png">
534 <assert_contents>
535 <has_image_width width="3223" delta="2"/>
536 <has_image_height height="1406" delta="2"/>
537 </assert_contents>
538 </output>
539 </test>
540 <test expect_num_outputs="2">
541 <param name="adata" value="tl.rank_genes_groups.newton-cg.pbmc68k_reduced_240cells.h5ad"/>
542 <param name="format" value="png"/>
543 <param name="probeset" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv"/>
544 <param name="header_probeset" value="included"/>
545 <param name="nreference" value="50"/>
546 <param name="methods_reference" value="random,PCA,DE,HVG"/>
547 <param name="genes_key" value="highly_variable"/>
548 <param name="obs_key" value="celltype"/>
549 <conditional name="method">
550 <param name="method" value="plot_marker_corr"/>
551 <param name="set_ids" value="all"/>
552 <param name="use_marker_corr" value="True"/>
553 <param name="markerset" value="marker.tsv"/>
554 <param name="header_markerset" value="not_included"/>
555 <param name="per_celltype" value="True"/>
556 <param name="per_marker" value="True"/>
557 <conditional name="select_per_celltype_min_mean">
558 <param name="use_per_celltype_min_mean" value="False"/>
559 </conditional>
560 <conditional name="select_per_marker_min_mean">
561 <param name="use_per_marker_min_mean" value="True"/>
562 <param name="per_marker_min_mean" value="0.025"/>
563 </conditional>
564 </conditional>
565 <param name="show_log" value="true" />
566 <output name="hidden_output">
567 <assert_contents>
568 <has_text_matching expression="'name': 'probeset',"/>
569 <has_text_matching expression="'per_celltype': True,"/>
570 <has_text_matching expression="'per_marker': True,"/>
571 <has_text_matching expression="'per_marker_min_mean': 0.025"/>
572 <has_text_matching expression="evaluator.plot_marker_corr"/>
573 <has_text_matching expression="save='plot.png'"/>
574 </assert_contents>
575 </output>
576 <output name="out_png">
577 <assert_contents>
578 <has_image_width width="5064" delta="2"/>
579 <has_image_height height="4554" delta="2"/>
580 </assert_contents>
581 </output>
582 </test>
583 </tests>
584 <help><![CDATA[
585 Probe set evaluation for single-cell sequencing data using spapros.
586 ============================================================================================================
587
588 Spapros is a python package that provides a pipeline for probe set selection and evaluation for targeted spatial transcriptomics data.
589
590 Key Features:
591 * Select probe sets for spatial transcriptomics which identify cell types of interest, capture general transcriptomic variation, and incorporate prior knowledge
592 * Evaluate probe sets with an extensive pipeline
593
594 Further documentation can be found here: https://spapros.readthedocs.io/en/latest/index.html.
595
596 ]]></help>
597 <expand macro="citations"/>
598 </tool>