add spacexr

tools/spacexr/.shed.yml

@@ -0,0 +1,13 @@
+name: spacexr
+owner: iuc
+description: Cell type identification and cell type-specific differential expression in spatial transcriptomics
+homepage_url: https://github.com/dmcable/spacexr/tree/master
+long_description: Computational methods for cell type identification (RCTD) and differential expression (C-SIDE) on spatial transcriptomics datasets
+remote_repository_url: https://github.com/galaxyproject/tools-iuc/tree/master/tools/spacexr
+categories: 
+- Spatial Analysis
+- Transcriptomics
+suite:
+  name: "suite_spacexr"
+  description: "A suite of Galaxy tools designed to work with the spacexr-tools collection."
+  type: repository_suite_definition

tools/spacexr/macros.xml

@@ -0,0 +1,31 @@
+<macros>
+    <token name="@TOOL_VERSION@">2.2.1</token>
+    <token name="@VERSION_SUFFIX@">0</token>
+    <token name="@PROFILE@">24.1</token>
+    <xml name="requirements">
+        <requirements>
+            <requirement type="package" version="@TOOL_VERSION@">r-spacexr</requirement>
+            <yield/>
+        </requirements>
+    </xml>
+    <xml name="edam">
+        <edam_topics>
+            <edam_topic>topic_4019</edam_topic>
+            <edam_topic>topic_4028</edam_topic>
+            <edam_topic>topic_3308</edam_topic>
+        </edam_topics>
+        <edam_operations>
+            <edam_operation>operation_3223</edam_operation>
+        </edam_operations>
+    </xml>            
+    <xml name="citations">
+        <citations>
+            <citation type="doi">10.1038/s41587-021-00830-w</citation>
+            <citation type="bibtex">@Manual{github,
+                title = {SpatialeXpressionR: Cell type identification and cell type-specific differential expression in spatial transcriptomics.},
+                author = {Dylan Cable},
+                url = {https://github.com/dmcable/spacexr}}
+            </citation>
+        </citations>
+    </xml>
+</macros>

tools/spacexr/spacexr_cside.xml

@@ -0,0 +1,26 @@
+<tool id="spacexr_cside" name="CSIDE" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
+    <description>Cell type-specific differential expression with C-SIDE</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="edam"/>
+    <expand macro="requirements"/>
+    <command detect_errors="exit_code"><![CDATA[
+
+    ]]></command>
+    <inputs>
+
+    </inputs>
+    <outputs>
+
+    </outputs>
+    <tests>
+
+    </tests>
+    <help><![CDATA[
+
+
+    ]]></help>
+
+    <expand macro="citations" />
+</tool>

tools/spacexr/spacexr_rctd.xml

@@ -0,0 +1,258 @@
+<tool id="spacexr_rctd" name="RCTD" version="@TOOL_VERSION@+galaxy@VERSION_SUFFIX@" profile="@PROFILE@">
+    <description>Cell type identification with RCTD</description>
+    <macros>
+        <import>macros.xml</import>
+    </macros>
+    <expand macro="edam"/>
+    <expand macro="requirements"/>
+    <command detect_errors="exit_code"><![CDATA[
+        mkdir -p 'results'
+        mkdir -p 'figures'
+
+    ]]></command>
+    <configfiles>
+        <configfile name="rctd_script">
+            # rctd script
+            # This file is used to specify the parameters for the rctd from spacexr package
+
+            # Load the spacexr library
+            library('spacexr')
+
+            ### Load the scRNA-seq data
+            counts &lt;- read.table(file = '${sc_count}', row.names = 1, sep = '\t')
+            metadata &lt;- read.table(file = '${metadata}', sep = '\t')
+
+            # create cell_types named list
+            cell_types &lt;- metadata[,"annotation"]; names(cell_types) &lt;- metadata[,"barcode"]
+
+            # convert to factor data type
+            cell_types &lt;- as.factor(cell_types)
+
+            #if str($sc_umi_input) == 'True':
+                # create nUMI named list
+                nUMI &lt;- meta_data[, "nUMI"]; names(nUMI) &lt;- meta_data[,"barcode"]
+            #end if
+
+            # Create reference object
+            reference &lt;- Reference(
+                                counts = counts,
+                                cell_types = cell_types,
+                                #if str($sc_umi_input) == 'True':
+                                    nUMI = nUMI,
+                                #end if
+                                n_max_cells = $n_max_cells,
+                                min_UMI = $min_UMI
+                                )
+
+            ### Load spatial data
+            counts &lt;- read.table(file = '${st_count}', row.names = 1, sep = '\t')
+            coords &lt;- read.table(file = '${coord}', row.names = 1, sep = '\t')
+
+            nUMI &lt;- colSums(counts) # In tutorials it is always the sum of counts
+
+            # Create SpatialRNA object
+            puck &lt;- SpatialRNA(
+                            coords = coords,
+                            counts = counts,
+                            nUMI= nUMI,
+                            )
+
+            # provide a basic plot of the nUMI of each pixel on the plot:
+            pdf('figures/nUMI_plot.pdf')
+            plot_puck_continuous(
+                puck = puck,
+                barcodes = colnames(puck@counts),
+                plot_val = puck@nUMI,
+                ylimit = c(0,round(quantile(puck@nUMI,0.9))),
+                title ='plot of nUMI')
+            dev.off()
+
+            ### Run the RCTD
+            myRCTD &lt;- create.RCTD(
+                            spatialRNA = puck,
+                            reference = reference,
+                            gene_cutoff = $gene_cutoff,
+                            fc_cutoff = $fc_cutoff,
+                            gene_cutoff_reg = $gene_cutoff_reg,
+                            fc_cutoff_reg = $fc_cutoff_reg,
+                            UMI_min = $UMI_min,
+                            UMI_max = $UMI_max,
+                            counts_MIN = $counts_MIN,
+                            UMI_min_sigma = $UMI_min_sigma,
+                            class_df = NULL, # set as default
+                            CELL_MIN_INSTANCE = $CELL_MIN_INSTANCE,
+                            #if str($cell_type_names) != "":
+                                cell_type_names = $cell_type_names,
+                            #end if                            
+                            MAX_MULTI_TYPES = $MAX_MULTI_TYPES,
+                            keep_reference = F, # set as default
+                            cell_type_profiles = NULL, # set as default
+                            CONFIDENCE_THRESHOLD = $CONFIDENCE_THRESHOLD,
+                            DOUBLET_THRESHOLD = $DOUBLET_THRESHOLD,)
+
+            myRCTD &lt;- run.RCTD(
+                            myRCTD,
+                            doublet_mode = $doublet_mode,)
+
+
+            # save results
+            #if str($doublet_mode) == 'doublet':
+                results &lt;- myRCTD@results
+
+                # save the data frame
+                result_df &lt;- results$results_df
+                write.table(result_df, file = 'results/doublet_results_df.tabular', sep = '\t', quote = F, row.names = T)
+
+                # RCTD plots
+                # normalize the cell type proportions to sum to 1.
+                norm_weights &lt;- normalize_weights(results$weights)
+                cell_type_names &lt;- myRCTD@cell_type_info$info[[2]] #list of cell type names
+                spatialRNA &lt;- myRCTD@spatialRNA
+
+                resultsdir &lt;- 'figures'
+
+                # make the plots
+                # Plots the confident weights for each cell type as in full_mode (saved as 'figures/cell_type_weights.pdf')
+                plot_weights(cell_type_names, spatialRNA, resultsdir, norm_weights) 
+
+                # Plots all weights for each cell type as in full_mode. (saved as 'figures/cell_type_weights_unthreshold.pdf')
+                plot_weights_unthreshold(cell_type_names, spatialRNA, resultsdir, norm_weights) 
+
+                # Plots the weights for each cell type as in doublet_mode. (saved as 'figures/cell_type_weights_doublets.pdf')
+                plot_weights_doublet(cell_type_names, spatialRNA, resultsdir, results$weights_doublet,results$results_df)
+
+                # Plots the number of confident pixels of each cell type in 'full_mode'. (saved as 'figures/cell_type_occur.pdf')
+                plot_cond_occur(cell_type_names, resultsdir, norm_weights, spatialRNA)
+
+                # makes a map of all cell types, (saved as 'results/all_cell_types.pdf')
+                plot_all_cell_types(results$results_df, spatialRNA@coords, cell_type_names, resultsdir) 
+
+                # doublets
+                #obtain a dataframe of only doublets
+                doublets &lt;- results$results_df[results$results_df$spot_class == "doublet_certain",] 
+
+                # Plots all doublets in space (saved as 'results/all_doublets.pdf')
+                plot_doublets(spatialRNA, doublets, resultsdir, cell_type_names)
+
+                # Plots all doublets in space for each cell type (saved as 'results/all_doublets_type.pdf')
+                plot_doublets_type(spatialRNA, doublets, resultsdir, cell_type_names) 
+
+                # a table of frequency of doublet pairs 
+                doub_occur &lt;- table(doublets$second_type, doublets$first_type) 
+                # Plots a stacked bar plot of doublet ocurrences (saved as 'results/doublet_stacked_bar.pdf')
+                plot_doub_occur_stack(doub_occur, resultsdir, cell_type_names) 
+
+                # save rds file
+                #if str($rds) == 'True':
+                    saveRDS(myRCTD, file = 'results/rctd_results_doublet.rds')
+                #end if
+            #end if
+
+            #if str($doublet_mode) == 'full':
+                results &lt;- myRCTD@results
+
+                # RCTD plots
+                # normalize the cell type proportions to sum to 1.
+                norm_weights &lt;- normalize_weights(results$weights)
+                cell_type_names &lt;- myRCTD@cell_type_info$info[[2]] #list of cell type names
+                spatialRNA &lt;- myRCTD@spatialRNA
+
+                resultsdir &lt;- 'figures'
+
+                # make the plots
+                # Plots the confident weights for each cell type as in full_mode (saved as 'figures/cell_type_weights.pdf')
+                plot_weights(cell_type_names, spatialRNA, resultsdir, norm_weights) 
+
+                # Plots all weights for each cell type as in full_mode. (saved as 'figures/cell_type_weights_unthreshold.pdf')
+                plot_weights_unthreshold(cell_type_names, spatialRNA, resultsdir, norm_weights) 
+
+                # Plots the number of confident pixels of each cell type in 'full_mode'. (saved as 'figures/cell_type_occur.pdf')
+                plot_cond_occur(cell_type_names, resultsdir, norm_weights, spatialRNA)
+
+                # save rds file
+                #if str($rds) == 'True':
+                    saveRDS(myRCTD, file = 'results/rctd_results_full.rds')
+                #end if
+            #end if
+        </configfile>
+    </configfiles>
+    <inputs>
+        <param name="sc_count" type="data" format="tabular" label="Single-cell count matrix" help="A matrix representing Digital Gene Expression (DGE). Rownames should be genes and colnames represent barcodes/cell names" />
+        <param name="metadata" type="data" format="tabular" label="Metadata" help="single-cell annotation file with columns: barcode, annotation, and nUMI(optional)" />
+        <param name="sc_umi_input" type="boolean" truevalue="True" falsevalue="False" checked="false" label="nUMI" help="Does your single-cell metadata have nUMI column?" />
+        <param name="st_count" type="data" format="tabular" label="Spatial count matrix" help="A matrix representing Digital Gene Expression (DGE). Rownames should be genes and colnames represent barcodes/pixel names" />
+        <param name="coord" type="data" format="tabular" label="Spatial coordinates" help="A numeric table representing the spatial pixel locations. Rownames are barcodes/pixel names, and there should be two columns for x and for y" />
+        <param name="doublet_mode" type="select" label="Doublet mode">
+            <option value="doublet">doublet</option>
+            <option value="full">full</option>
+        </param>
+        <section name="advanced_param" title="Advanced parameters">
+            <section name="reference_param" title="Reference object parameters">
+                <param name="n_max_cells" type="integer" min="0" value="10000" label="n_max_cells" help="Maximum number of cells per cell type. Will downsample if this number is exceeded." />
+                <param name="min_UMI" type="integer" min="0" value="100" label="min_UMI" help="Minimum UMI count for cells to be included in the reference." />
+            </section>
+            <section name="st_param" title="SpatialRNA object parameters">
+                <param name="gene_cutoff" type="float" min="0" value="0.000125" label="gene_cutoff" help="Minimum normalized gene expression for genes to be included in the platform effect normalization step." />
+                <param name="fc_cutoff" type="float" min="0" value="0.5" label="fc_cutoff" help="Minimum log-fold-change (across cell types) for genes to be included in the platform effect normalization step." />
+                <param name="gene_cutoff_reg" type="float" min="0" value="0.0002" label="gene_cutoff_reg" help="Minimum normalized gene expression for genes to be included in the RCTD step." />
+                <param name="fc_cutoff_reg" type="float" min="0" value="0.75" label="fc_cutoff_reg" help="Minimum log-fold-change (across cell types) for genes to be included in the RCTD step." />
+                <param name="UMI_min" type="integer" min="0" value="100" label="UMI_min" help="Minimum UMI per pixel included in the analysis" />
+                <param name="UMI_max" type="integer" min="0" value="20000000" label="UMI_max" help="Maximum UMI per pixel included in the analysis" />
+                <param name="counts_MIN" type="integer" min="0" value="10" label="counts_MIN" help="Minimum total counts per pixel of genes used in the analysis." />
+                <param name="UMI_min_sigma" type="integer" min="0" value="300" label="UMI_min_sigma" help="Minimum UMI per pixel for the choose_sigma_c function" />
+                <param name="CELL_MIN_INSTANCE" type="integer" min="0" value="25" label="CELL_MIN_INSTANCE" help="Minimum number of cells required per cell type." />
+                <param name="cell_type_names" type="text" label="cell_type_names" help="A list of cell types to be included from the reference. If NULL, uses all cell types" />
+                <param name="MAX_MULTI_TYPES" type="integer" min="0" value="4" label="MAX_MULTI_TYPES" help="Max number of cell types per pixel." />
+                <param name="CONFIDENCE_THRESHOLD" type="integer" min="0" value="5" label="CONFIDENCE_THRESHOLD" help="The minimum change in likelihood (compared to other cell types) necessary to determine a cell type identity with confidence." />
+                <param name="DOUBLET_THRESHOLD" type="integer" min="0" value="20" label="DOUBLET_THRESHOLD" help="The penalty weight of predicting a doublet instead of a singlet for a pixel." />
+            </section>
+        </section>
+        <section name="output" title="Output options">
+            <param name="rds" type="boolean" truevalue="True" falsevalue="False" checked="false" label="save RDS file?"/>
+        </section>
+
+
+
+
+
+
+
+
+
+
+
+    </inputs>
+    <outputs>
+
+    </outputs>
+    <tests>
+
+    </tests>
+    <help><![CDATA[
+
+Robust Cell Type Decomposition, or **RCTD**, is an statistical method for learning cell types from spatial transcriptomics data.
+
+**Reference object**:
+
+To create the single-cell reference object, the following inputs are required:
+    * counts: A matrix representing Digital Gene Expression (DGE). **Rownames** should be genes and **colnames** represent barcodes/cell names. Counts should be **untransformed** count-level data.
+    * cell_types: A named (by cell barcode) factor of cell type for each cell.
+    * nUMI: Optional, a named (by cell barcode) list of total counts or UMI&apos;s appearing at each cell. If not provided, nUMI will be assumed to be the total counts appearing on each cell. 
+
+**SpatialRNA object**:
+
+To create the spatialRNA object, the following inputs are required:
+    * coords: A numeric table representing the spatial pixel locations. **Rownames** are barcodes/pixel names, and there should be two columns for **x** and for **y**.
+    * counts: A matrix representing Digital Gene Expression (DGE). **Rownames** should be genes and **colnames** represent barcodes/pixel names. Counts should be **untransformed** count-level data.
+
+-----
+
+RCTD has **three** modes:
+    * **doublet mode**, which assigns 1-2 cell types per spot and is recommended for technologies with high spatial resolution such as Slide-seq and MERFISH.
+    * **full mode**, which assigns any number of cell types per spot and is recommended for technologies with poor spatial resolution such as 100-micron resolution Visium.
+    * **multi mode**, an extension of doublet mode that can discover more than two cell types per spot (3-4 cell types) as an alternative option to full mode.
+
+    ]]></help>
+
+    <expand macro="citations" />
+</tool>