feat: enhance analysis and visualization capabilities for scRNA-seq data

- Update service description for comprehensive single-cell RNA sequencing analysis.
- Add new input parameters: max_features, resolution, species, min_pct, logfc_threshold, gsea_min_size, gsea_max_size, and category.
- Implement GF-ICF for single-cell GSEA and integrate pathway analysis.
- Enhance output files to include detailed cluster information and pathway scores.
- Introduce docker-compose setup for RStudio environment.
- Optimize Dockerfile with streamlined package installations using `pak`.

.osparc/metadata.yml

@@ -3,7 +3,8 @@ key: simcore/services/comp/osparc-differential-expression
 type: computational
 integration-version: 1.0.0
 version: 0.1.0
-description: Easily generate differential expression results from OSparc data
+description: |
+  Easily generate differential expression results from OSparc data. This service performs comprehensive analysis of single-cell RNA sequencing data, including data normalization, clustering, dimensionality reduction, and pathway activity scoring. It provides visualizations such as t-SNE and UMAP plots for both gene expression and pathway activity, along with detailed cluster information and pathway scores. The service is designed to work with standard input formats and offers flexibility in analysis parameters.
 contact: [email protected]
 thumbnail: https://github.com/ITISFoundation/osparc-assets/blob/cb43207b6be2f4311c93cd963538d5718b41a023/assets/default-thumbnail-cookiecutter-osparc-service.png?raw=true
 authors:
@@ -16,11 +17,12 @@ inputs:
     label: Input Folder
     description: Folder containing scRNA-seq data (matrix.mtx, features.tsv, barcodes.tsv)
     type: data:*/*
-  output_prefix:
+  name:
     displayOrder: 2
-    label: Output Prefix
-    description: Prefix for output files (optional)
+    label: Project Name
+    description: The name of the dataset being analyzed
     type: string
+    defaultValue: sPARcRNA
   min_cells:
     displayOrder: 3
     label: Minimum Cells
@@ -33,11 +35,73 @@ inputs:
     description: Minimum number of features (genes) per cell
     type: integer
     defaultValue: 200
+  max_features:
+    displayOrder: 5
+    label: Maximum Features
+    description: Maximum number of features (genes) per cell
+    type: integer
+    defaultValue: 2500
+  resolution:
+    displayOrder: 6
+    label: Resolution
+    description: Resolution parameter for clustering
+    type: number
+    defaultValue: 0.8
+  species:
+    displayOrder: 7
+    label: Species
+    description: Species for GSEA (e.g., "Homo sapiens" or "Mus musculus")
+    type: string
+    defaultValue: Homo sapiens
+  min_pct:
+    displayOrder: 8
+    label: Minimum Percentage
+    description: Minimum percentage for FindAllMarkers
+    type: number
+    defaultValue: 0.25
+  logfc_threshold:
+    displayOrder: 9
+    label: Log Fold-Change Threshold
+    description: Log fold-change threshold for FindAllMarkers
+    type: number
+    defaultValue: 0.25
+  gsea_min_size:
+    displayOrder: 10
+    label: GSEA Minimum Size
+    description: Minimum gene set size for GSEA
+    type: integer
+    defaultValue: 15
+  gsea_max_size:
+    displayOrder: 11
+    label: GSEA Maximum Size
+    description: Maximum gene set size for GSEA
+    type: integer
+    defaultValue: 500
+  category:
+    displayOrder: 12
+    label: MSigDB Category
+    description: MSigDB category for GSEA (e.g., "H" for hallmark gene sets)
+    type: string
+    defaultValue: H
 outputs:
   output_file:
     displayOrder: 1
     label: Processed Data
-    description: Zipped file containing processed scRNA-seq data (AnnData objects) and metadata
+    description: |
+      Zipped file containing processed scRNA-seq data and analysis results. The zip file includes:
+      - seurat_object.rds: Serialized R object containing the complete Seurat analysis
+      - tsne_plot.png: t-SNE plot of cell clusters based on gene expression
+      - dim_reduction_data.csv: CSV file with UMAP and t-SNE coordinates for both gene expression and pathway activity, along with cluster assignments
+      - pathway_scores.csv: CSV file with pathway activity scores for each cell
+      - cluster_info_genes.csv: CSV file with cluster information based on gene expression, including centroids and top pathways
+      - cluster_info_pathways.csv: CSV file with cluster information based on pathway activity, including centroids and top pathways
+      - outputs.json: JSON file containing summary statistics and file paths, including:
+        - initial_cell_count: Number of cells in the input data
+        - final_cell_count: Number of cells after filtering
+        - gene_count: Number of genes in the analysis
+        - project_name: Name of the analyzed dataset
+        - cluster_count_genes: Number of clusters based on gene expression
+        - cluster_count_pathways: Number of clusters based on pathway activity
     type: data:application/zip
     fileToKeyMap:
       final_output.zip: output_file

docker-compose-rstudio.yaml

@@ -0,0 +1,18 @@
+services:
+  rstudio:
+    image: rocker/rstudio:4.2.0
+    container_name: rstudio_instance
+    environment:
+      - PASSWORD=yourpassword
+      - USER=scu
+      - INPUT_FOLDER=/input
+      - OUTPUT_FOLDER=/output
+    volumes:
+      - ./src/pipeline:/home/scu/pipeline
+      - ./src/astro:/home/scu/astro
+      - ./input:/input
+      - ./output:/output
+    ports:
+      - "8787:8787"
+    user: "root"
+    command: ["--server-user", "scu"]

docker/Dockerfile

@@ -12,7 +12,7 @@ ENV NODE_VERSION="22.x" \
 # Install system packages and R
 RUN apt-get update \
   && apt-get -y install --no-install-recommends \
-    curl jq zip adduser apt-transport-https ca-certificates gnupg \
+    curl jq zip adduser apt-transport-https ca-certificates gnupg libgsl-dev \
     libcurl4-openssl-dev libssl-dev libxml2-dev libhdf5-dev python3 python3-pip \
     libfontconfig1-dev libharfbuzz-dev libfribidi-dev libfreetype6-dev libpng-dev libtiff5-dev libjpeg-dev \
   && echo "deb https://cloud.r-project.org/bin/linux/ubuntu noble-cran40/" >> /etc/apt/sources.list \
@@ -38,13 +38,9 @@ RUN adduser --uid ${SC_USER_ID} --disabled-password --gecos "" --shell /bin/sh -
 
 # ------------------------------------------------------------------------------------
 # Install Bioconductor
-RUN R -e "install.packages('BiocManager', repos='http://cran.rstudio.com/')"
-# Install required R packages available on CRAN
-RUN R -e "install.packages(c('Seurat', 'jsonlite', 'optparse', 'ggplot2', 'devtools', 'dplyr', 'msigdbr'), repos='http://cran.rstudio.com/')"
-# Install required R packages available on Bioconductor
-RUN R -e "BiocManager::install('fgsea')"
-# Install presto for speed-ups to Seurat
-RUN R -e "library(devtools); devtools::install_github('immunogenomics/presto')"
+
+RUN R -e "install.packages(c('pak', 'devtools', 'BiocManager'), repos='http://cran.rstudio.com/')"
+RUN R -e "pak::pkg_install(c('Seurat', 'jsonlite', 'optparse', 'ggplot2', 'devtools', 'dplyr', 'msigdbr', 'BiocManager', 'sva', 'edgeR', 'fgsea', 'immunogenomics/presto', 'zdebruine/RcppML', 'gambalab/gficf'))"
 
 # Remove the Python warning about system installs since we're in a docker image
 RUN rm /usr/lib/python*/EXTERNALLY-MANAGED