Mercurial > repos > iuc > scanpy_filter

changeset 15:aa0059118fb9 draft

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/scanpy/ commit c21958f44b81d740191999fb6015d5ae69538ee0
author iuc
date Wed, 31 Jul 2024 18:10:52 +0000
parents d636ce5cde16
children 72a6bebab2a5
files filter.xml macros.xml test-data/cosg.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_1.h5ad test-data/tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_1.tsv test-data/tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv
diffstat 5 files changed, 130 insertions(+), 5 deletions(-) [+]
line wrap: on
line diff
--- a/filter.xml	Sat May 18 18:29:27 2024 +0000
+++ b/filter.xml	Wed Jul 31 18:10:52 2024 +0000
@@ -98,6 +98,41 @@
     random_state=$method.random_state,
     replace=$method.replace,
     copy=False)
+
+#else if $method.method == "filter_marker"
+
+#if $method.layer_selection.use_raw == 'False':
+    adata.X = adata.layers['$method.layer_selection.layer']
+#end if
+
+def check_marker(adata, group, gene, thresh_mean, thresh_frac, groupby):
+    filtered_data = adata[adata.obs[groupby] == group, adata.var_names == gene]
+    mean_expression = np.mean(filtered_data.X)
+    frac_cell_mean_expression = len(filtered_data.X[filtered_data.X > mean_expression]) / filtered_data.n_obs
+    if ( mean_expression > thresh_mean and frac_cell_mean_expression >= thresh_frac ):
+        return(True)
+    return(False)
+
+header='infer'
+
+#if $method.header == 'not_included':
+    header=None
+#end if
+
+marker_list={key: list(value.values()) for key, value in pd.read_csv('$method.markerfile', sep='\t', index_col=0, header=header).to_dict(orient='index').items()}
+
+for key, value in marker_list.items():
+    marker_list[key] = [x for x in value if check_marker(adata, key, x, $method.thresh_mean, $method.thresh_frac, '$method.groupby')]
+
+# Find the maximum length of lists
+max_len = max(len(lst) for lst in marker_list.values())
+
+# Fill smaller lists with empty values
+for key, value in marker_list.items():
+    marker_list[key] = value + [''] * (max_len - len(value))
+
+df = pd.DataFrame(marker_list).T
+df.to_csv('marker.tsv', sep='\t', index=True)
 #end if
 
 @CMD_anndata_write_outputs@
@@ -113,6 +148,7 @@
                 <option value="pp.highly_variable_genes">Annotate (and filter) highly variable genes, using 'pp.highly_variable_genes'</option>
                 <option value="pp.subsample">Subsample to a fraction of the number of observations, using 'pp.subsample'</option>
                 <option value="pp.downsample_counts">Downsample counts from count matrix, using 'pp.downsample_counts'</option>
+                <option value="filter_marker">Filter markers from count matrix and marker list</option>
             </param>
             <when value="pp.filter_cells">
                 <conditional name="filter">
@@ -213,11 +249,36 @@
                 <param argument="random_state" type="integer" value="0" label="Random seed to change subsampling"/>
                 <param argument="replace" type="boolean" truevalue="True" falsevalue="False" checked="false" label="Sample the counts with replacement?"/>
             </when>
+            <when value="filter_marker">
+                <param argument="markerfile" type="data" format="tabular" label="List of markers" help="This should be a tsv where row = group (e.g. celltypes) and columns = markers."></param>
+                <param name="header" type="select" label="Header in the list of markers?">
+                    <option value="included">Header incldued</option>
+                    <option value="not_included">Header not included</option>
+                </param>
+                <param argument="thresh_mean" type="float" min="0.0" value="1.0" label="Minimal average count of all cells of a group (e.g., celltype) for a particular marker" help="Increasing the threshold will result in a smaller marker set."/>
+                <param argument="thresh_frac" type="float" min="0.0" max="1.0" value="0.1" label="Minimal fractions of cells that has a higher count than the average count of all cells of the group for the marker" help="Increasing this threshold might remove marker outliers."/>
+                <conditional name="layer_selection">
+                    <param name="use_raw" type="select" label="Use .X of adata to perform the filtering" help="">
+                        <option value="True">Yes</option>
+                        <option value="False">No</option>
+                    </param>
+                    <when value="False">
+                        <param argument="layer" type="text" value="" label="Key from adata.layers whose value will be used to filter" help="If layers specified then use adata.layers[layer]."/>
+                    </when>
+                    <when value="True"/>
+                </conditional>
+                <param argument="groupby" type="text" value="" label="The key of the observation grouping to consider (e.g., celltype)" help="">
+                    <expand macro="sanitize_query" />
+                </param>
+            </when>
         </conditional>
         <expand macro="inputs_common_advanced"/>
     </inputs>
     <outputs>
         <expand macro="anndata_outputs"/>
+        <data name="marker_out" format="tabular" from_work_dir="marker.tsv" label="${tool.name} on ${on_string}: Markers">
+            <filter>method['method'] == 'filter_marker'</filter>
+        </data>
     </outputs>
     <tests>
         <test expect_num_outputs="2">
@@ -444,6 +505,32 @@
             </output>
             <output name="anndata_out" file="pp.downsample_counts.random-randint.h5ad" ftype="h5ad" compare="sim_size" delta="10000000" delta_frac="0.5"/>
         </test>
+        <test expect_num_outputs="3">
+            <!-- test 10 -->
+            <param name="adata" value="cosg.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_1.h5ad" />
+            <conditional name="method">
+                <param name="method" value="filter_marker"/>
+                <param name="markerfile" value="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_1.tsv"/>
+                <param name="thresh_mean" value="1.0"/>
+                <param name="thresh_frac" value="0.2"/>
+                <param name="layer_selection" value="True"/>
+                <param name="groupby" value="bulk_labels"/>
+            </conditional>
+            <section name="advanced_common">
+                <param name="show_log" value="true" />
+            </section>
+            <output name="hidden_output">
+                <assert_contents>
+                    <has_text_matching expression="adata, key, x, 1.0, 0.2, 'bulk_labels'"/>
+                </assert_contents>
+            </output>
+            <output name="anndata_out" file="cosg.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_1.h5ad" ftype="h5ad">
+                <assert_contents>
+                    <has_h5_keys keys="obs, var, uns" />
+                </assert_contents>
+            </output>
+            <output name="marker_out" file="tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv" ftype="tabular" compare="sim_size"/>
+        </test>
     </tests>
     <help><![CDATA[
 
@@ -502,6 +589,16 @@
 Downsample counts so that each cell has no more than `target_counts`. Cells with fewer counts than `target_counts` are unaffected by this. This
 has been implemented by M. D. Luecken.
 
+
+Filter marker genes (`filter_marker`)
+======================================================================
+
+This option is specific for celltype marker gene detection. You can generate a celltype marker gene file (tsv) with **COSG** provided at Galaxy.
+
+The marker gene file should have as rows celltypes and columns as marker genes. Each celltype can have varying number of marker genes.
+
+A marker gene is returned (retained in the list) if the mean expression of the marker gene is bigger than the threshold of mean expression (thresh_mean) and if the fraction of cells with the marker gene expression is equal or higher than the cell fraction threshold (thresh_frac).
+
 More details on the `scanpy documentation
 <https://scanpy.readthedocs.io/en/stable/api/scanpy.pp.downsample_counts.html>`__
 
--- a/macros.xml	Sat May 18 18:29:27 2024 +0000
+++ b/macros.xml	Wed Jul 31 18:10:52 2024 +0000
@@ -1,6 +1,6 @@
 <macros>
     <token name="@TOOL_VERSION@">1.9.6</token>
-    <token name="@VERSION_SUFFIX@">3</token>
+    <token name="@VERSION_SUFFIX@">4</token>
     <token name="@profile@">21.09</token>
     <xml name="requirements">
         <requirements>
@@ -425,6 +425,7 @@
             <param name="type" type="select" label="Variables to plot (columns of the heatmaps)" >
                 <option value="all">All variables in 'adata.var_names'</option>
                 <option value="custom">Subset of variables in 'adata.var_names'</option>
+                <option value="customfile">Subset of variables as a tabular file</option>
             </param>
             <when value="all"/>
             <when value="custom">
@@ -432,6 +433,9 @@
                     <expand macro="sanitize_query" />
                 </param>
             </when>
+            <when value="customfile">
+                <param argument="var_names" type="data" format="tabular" label="List of variables to plot" help="This should be a tsv where row = group (e.g. celltypes) and columns = variables."></param>
+            </when>
         </conditional>
     </xml>
     <xml name="param_num_categories">
@@ -468,11 +472,13 @@
         <expand macro="param_num_categories"/>
     </xml>
     <token name="@CMD_params_inputs@"><![CDATA[
-    #if $method.var_names.type == 'all'
-    var_names=adata.var_names,
-    #else
+    #if $method.var_names.type == 'custom'
     #set $var_names = ([x.strip() for x in str($method.var_names.var_names).split(',')])
     var_names=$var_names,
+    #else if $method.var_names.type == 'customfile'
+    var_names={key: [v for v in list(value.values()) if pd.notna(v)] for key, value in pd.read_csv('$var_names', sep='\t', index_col=0).to_dict(orient='index').items()},
+    #else
+    var_names=adata.var_names,
     #end if
     #if $method.groupby
     groupby='$method.groupby',
@@ -1122,7 +1128,7 @@
         </param>
     </xml>
     <xml name="param_n_genes">
-        <param argument="n_genes" type="integer" min="0" value="10" label="Number of genes to show" help=""/>
+        <param argument="n_genes" type="integer" min="0" value="10" label="Number of genes to show" help="It is only used if you are not specifying certain variable names"/>
     </xml>
     <xml name="pl_dotplot">
         <param argument="color_map" type="select" optional="true" label="Color palette">
Binary file test-data/cosg.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_1.h5ad has changed
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_1.tsv	Wed Jul 31 18:10:52 2024 +0000
@@ -0,0 +1,11 @@
+	0	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25	26	27	28	29	30	31	32	33	34	35	36	37	38	39	40	41	42	43	44	45	46	47	48	49
+CD14+ Monocyte	PILRA	PSAP	CD68	TMEM176B	FTL	NPC2	LST1	FCGR3A	FCER1G	CEBPB	FCN1	SERPINA1	OAZ1	CFD	FTH1	HCK	AIF1	SAT1	CTSS	S100A11	MS4A7	TYROBP	COTL1	STXBP2	RP11-290F20.3	S100A4	IFITM2	SPI1	DUSP1	SESN2	IFITM3	MPP1	GALE	CORO1B	RP11-390E23.6	VIMP	RSBN1L-AS1	CHD4	CFP	GSTP1	PFN1	FCGRT	ADTRP	ARHGDIB	AMICA1	HLA-DRB5	CST3	GRN	HLA-DPA1	SSR3
+CD19+ B	TNFRSF13B	CD79B	SMARCB1	PNOC	CCDC50	AL928768.3	BANK1	MS4A1	CD79A	ISG20	IGLL5	TNFRSF17	KIAA0125	TPD52	PEBP1	FKBP11	CCDC132	SUB1	POU2AF1	MZB1	PTPRCAP	UBE2J1	BLK	SPIB	DERL3	FAM63B	MPHOSPH9	IGJ	FCRLA	XBP1	NCF1	SSR3	CD52	TSHZ2	PDLIM1	VIMP	SSR4	S1PR4	SELL	HMGA1	NUCB2	JUN	CD27	ARHGDIB	GYPC	CALR	ADTRP	BTG1	EXOG	RARRES3
+CD34+	PRSS57	C19orf77	SPINK2	RP11-620J15.3	SNHG7	CYTL1	EGFL7	NGFRAP1	SOX4	NFE2	EGR1	RP3-467N11.1	H1FX	CDK6	SERPINB1	SPINT2	HMGA1	IL1B	NUCB2	RPLP0	IGFBP7	RPLP1	ATXN7L3B	RPS3	C1orf228	KIAA0125	RPL3	SYPL1	CD63	LDHB	SEPT1	JUN	FAM101B	PRKCQ-AS1	MATK	PEBP1	SELL	ITM2A	SSR3	SPON2	XBP1	UBE2J1	VIMP	GYPC	STK17A	STMN1	VIM	MZB1	HOPX	CD99
+CD4+/CD25 T Reg	IL32	SPOCK2	ACTG1	CD2	CD3D	GPR171	ARHGDIB	ACOX1	MAL	SIT1	GIMAP4	AES	CD52	SEPT1	TMSB10	LAT	STMN1	LINC00402	CD27	TSHZ2	S1PR4	CD3E	PFN1	CD99	AQP3	PTPRCAP	CD3G	LY9	LCK	CD247	S100A4	CCR7	TTC39C	CORO1B	MPHOSPH9	FYB	RPSA	FLT3LG	B2M	GIMAP7	PRKCQ-AS1	SELL	BTG1	CCDC132	GYPC	DENND2D	LDHB	IL7R	ITM2A	RPLP0
+CD4+/CD45RA+/CD25- Naive T	EAF2	GNG7	SSR4	CALR	DERL3	MANF	IGJ	XBP1	ATXN7L3B	SSR3	UBE2J1	CD79A	MZB1	RP3-467N11.1	TNFRSF17	NCF1	CDK6	SUB1	POU2AF1	AL928768.3	FKBP11	VIMP	GYPC	JUN	CD27	PEBP1	SMARCB1	FLT3LG	RPLP1	RPLP0	CCDC50	ISG20	IGLL5	HCST	GSTP1	GPX1	CD52	VIM	PTPRCAP	FCGRT	CD74	B2M	RPL3	CYTL1	SPINK2	PRSS57	C19orf77	RP11-620J15.3	FAM101B	CCDC132
+CD4+/CD45RO+ Memory	RNF138	NOSIP	IFITM1	LCK	RARRES3	ALOX5AP	FAM63B	RAB3IP	GZMK	CD3G	SEPT1	LDHB	SELL	CD3D	EXOG	RPSA	CD247	AES	CD52	TMSB10	NUCB2	DENND2D	RPL3	RPLP1	ACTG1	FYB	GIMAP7	CORO1B	LY9	CD7	PFN1	RPS3	GYPC	CD2	ARHGDIB	IL32	RPLP0	CD99	CD3E	GIMAP4	HCST	B2M	LAT	ISG20	ITM2A	FKBP11	SERPINB1	STK17A	CCR7	PTPRCAP
+CD56+ NK	CST7	SPON2	HOPX	GNLY	NKG7	CTSW	KLRC2	CD7	MATK	PCIF1	CLIC3	FGFBP2	SYPL1	GZMB	C9orf142	PRF1	CD247	HCST	GZMA	GZMH	STMN1	ALOX5AP	CD63	CD99	IGFBP7	GZMM	CCL5	B2M	DENND2D	GIMAP7	RARRES3	SIT1	IFITM1	PFN1	EXOG	XBP1	IFITM2	GIMAP4	VIMP	STK17A	LCK	GZMK	SEPT1	SSR3	CD8A	CD3G	SPOCK2	RPS3	LDHB	IL32
+CD8+ Cytotoxic T	FAM101B	ADTRP	GZMK	HCST	LAT	EGR1	CD8B	CCL5	RPL3	LINC00402	FGFBP2	GZMM	RPS3	CD3E	GYPC	DENND2D	C9orf142	GZMA	SEPT1	JUN	FYB	CD8A	SELL	ALOX5AP	CD3G	STK17A	AQP3	C1orf228	CD3D	HOPX	NKG7	CD2	NGFRAP1	RPLP1	RPSA	CCR7	IL7R	SPON2	PRF1	RARRES3	PRKCQ-AS1	FKBP11	MANF	CTSW	GNLY	CD27	LDHB	MAL	LTB	RPLP0
+CD8+/CD45RA+ Naive Cytotoxic	RP11-291B21.2	CD8A	CD8B	RSBN1L-AS1	GIMAP5	GZMM	GALE	CCR7	STK17A	RAB3IP	GZMH	GIMAP7	CD3E	C1orf228	LCK	CCL5	PEBP1	CD27	GYPC	LDHB	RNF34	CD99	CD3G	PFN1	IL7R	CD2	C9orf142	TMSB10	NGFRAP1	S1PR4	ITM2A	CD7	RPS3	IL32	FYB	IFITM1	CD52	LAT	GIMAP4	MAL	STMN1	NOSIP	RARRES3	SPOCK2	ACTG1	PRF1	CD3D	RPLP1	SELL	GZMA
+Dendritic	HLA-DQB1	CST3	HLA-DRB1	HLA-DQA2	HLA-DQA1	LYZ	HLA-DPB1	HLA-DPA1	HLA-DMA	HLA-DRA	VIM	CD74	ALDH2	FCER1A	GPX1	HLA-DRB5	LGALS2	MNDA	FCGRT	GRN	HLA-DMB	FOS	CPVL	CLEC10A	AMICA1	CFP	LY86	GSTP1	RP11-473M20.7	IL1B	GSN	SPINT2	CCDC163P	IGFBP7	EXOG	DUSP1	CD63	COTL1	FTH1	SPI1	TYROBP	SPIB	S100A11	OAZ1	CTSS	CCDC50	AIF1	SERPINB1	TMSB10	PCIF1
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/test-data/tl.rank_genes_groups.newton-cg.pbmc68k_highly_reduced_marker_filtered_1.tsv	Wed Jul 31 18:10:52 2024 +0000
@@ -0,0 +1,11 @@
+	0	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25	26	27	28	29	30	31	32	33
+CD14+ Monocyte	PILRA	PSAP	CD68	TMEM176B	FTL	NPC2	LST1	FCGR3A	FCER1G	CEBPB	FCN1	SERPINA1	OAZ1	CFD	FTH1	HCK	AIF1	SAT1	CTSS	S100A11	TYROBP	COTL1	S100A4	SPI1	DUSP1									
+CD19+ B	TNFRSF13B	CD79B	SMARCB1	PNOC	CCDC50	AL928768.3	BANK1	MS4A1	CD79A	ISG20	IGLL5	TNFRSF17	KIAA0125	TPD52	PEBP1	FKBP11	SUB1	POU2AF1	MZB1	PTPRCAP	DERL3	XBP1	CD52											
+CD34+	PRSS57	C19orf77	SPINK2	RP11-620J15.3	SNHG7	CYTL1	EGFL7	NGFRAP1	SOX4	NFE2	EGR1	RP3-467N11.1	H1FX	CDK6	SERPINB1	SPINT2	HMGA1	IL1B	NUCB2	RPLP0	IGFBP7	RPLP1	ATXN7L3B	RPS3	KIAA0125	RPL3	SYPL1	LDHB	SEPT1	FAM101B				
+CD4+/CD25 T Reg	IL32	SPOCK2	ACTG1	CD2	CD3D	ARHGDIB	GIMAP4	AES	CD52	SEPT1	TMSB10	LAT	STMN1	CD27	CD3E	PFN1	CD99	PTPRCAP	CD3G	LCK	CD247													
+CD4+/CD45RA+/CD25- Naive T																																		
+CD4+/CD45RO+ Memory	RNF138	NOSIP	IFITM1	LCK	RARRES3	ALOX5AP	RAB3IP	GZMK	CD3G	SEPT1	LDHB	SELL	CD3D	EXOG	RPSA	CD247	AES	CD52	TMSB10	RPL3														
+CD56+ NK	CST7	SPON2	HOPX	GNLY	NKG7	CTSW	KLRC2	CD7	MATK	PCIF1	CLIC3	FGFBP2	SYPL1	GZMB	C9orf142	PRF1	CD247	HCST	GZMA	GZMH	STMN1	ALOX5AP	CD63	CD99	IGFBP7	GZMM	CCL5	B2M	DENND2D	GIMAP7	RARRES3	SIT1	PFN1	XBP1
+CD8+ Cytotoxic T	FAM101B	ADTRP	GZMK	HCST	LAT	EGR1	CD8B	CCL5	RPL3	LINC00402	FGFBP2	GZMM	RPS3	CD3E	GYPC	C9orf142	SEPT1	SELL	CD3G	CD3D	HOPX	NGFRAP1												
+CD8+/CD45RA+ Naive Cytotoxic	RP11-291B21.2	CD8A	CD8B	RSBN1L-AS1	GIMAP5	GZMM	GALE	CCR7	STK17A	RAB3IP	GIMAP7	CD3E	LCK	PEBP1	CD27	LDHB	CD99	CD3G	C9orf142	NGFRAP1	IL32													
+Dendritic	HLA-DQB1	CST3	HLA-DRB1	HLA-DQA2	HLA-DQA1	HLA-DPB1	HLA-DPA1	HLA-DRA	CD74