Initiation of new Glioblastoma analysis module

.github/workflows/docker_celltype-glioblastoma.yml

@@ -0,0 +1,63 @@
+# This is a workflow to build the docker image for the celltype-glioblastoma module
+#
+# Docker modules are run on pull requests when code for files that affect the Docker image have changed.
+# If other files are used during the Docker build, they should be added to `paths`
+#
+# At module initialization, this workflow is inactive, and needs to be activated manually
+
+name: Build docker image for celltype-glioblastoma
+
+concurrency:
+  # only one run per branch at a time
+  group: "docker_celltype-glioblastoma_${{ github.ref }}"
+  cancel-in-progress: true
+
+on:
+  # pull_request:
+  #   branches:
+  #     - main
+  #   paths:
+  #     - "analyses/celltype-glioblastoma/Dockerfile"
+  #     - "analyses/celltype-glioblastoma/.dockerignore"
+  #     - "analyses/celltype-glioblastoma/renv.lock"
+  #     - "analyses/celltype-glioblastoma/conda-lock.yml"
+  # push:
+  #   branches:
+  #     - main
+  #   paths:
+  #     - "analyses/celltype-glioblastoma/Dockerfile"
+  #     - "analyses/celltype-glioblastoma/.dockerignore"
+  #     - "analyses/celltype-glioblastoma/renv.lock"
+  #     - "analyses/celltype-glioblastoma/conda-lock.yml"
+  workflow_dispatch:
+    inputs:
+      push-ecr:
+        description: "Push to AWS ECR"
+        type: boolean
+        required: true
+
+jobs:
+  test-build:
+    name: Test Build Docker Image
+    if: github.event_name == 'pull_request' || (contains(github.event_name, 'workflow_') && !inputs.push-ecr)
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Build image
+        uses: docker/build-push-action@v5
+        with:
+          context: "{{defaultContext}}:analyses/celltype-glioblastoma"
+          push: false
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+
+  build-push:
+    name: Build and Push Docker Image
+    if: github.repository_owner == 'AlexsLemonade' && (github.event_name == 'push' || inputs.push-ecr)
+    uses: ./.github/workflows/build-push-docker-module.yml
+    with:
+      module: "celltype-glioblastoma"
+      push-ecr: true

.github/workflows/run_celltype-glioblastoma.yml

@@ -0,0 +1,62 @@
+# This is a workflow to run the celltype-glioblastoma module
+#
+# Analysis modules are run based on three triggers:
+# - Manual trigger
+# - On pull requests where code in the module has changed
+# - As a reusable workflow called from a separate workflow which periodically runs all modules
+#
+# At initialization, only the manual trigger is active
+
+name: Run celltype-glioblastoma analysis module
+env:
+  MODULE_PATH: analyses/celltype-glioblastoma
+  AWS_DEFAULT_REGION: us-east-2
+
+concurrency:
+  # only one run per branch at a time
+  group: "run_celltype-glioblastoma_${{ github.ref }}"
+  cancel-in-progress: true
+
+on:
+  workflow_dispatch:
+  # workflow_call:
+  # pull_request:
+  #   branches:
+  #     - main
+  #   paths:
+  #     - analyses/celltype-glioblastoma/**
+  #     - "!analyses/celltype-glioblastoma/Dockerfile"
+  #     - "!analyses/celltype-glioblastoma/.dockerignore"
+  #     - .github/workflows/run_celltype-glioblastoma.yml
+
+jobs:
+  run-module:
+    if: github.repository_owner == 'AlexsLemonade'
+    runs-on: ubuntu-latest
+    defaults:
+      run:
+        shell: bash -el {0}
+
+    steps:
+      - name: Checkout repo
+        uses: actions/checkout@v4
+
+      - name: Set up conda
+        # Note that this creates and activates an environment named 'test' by default
+        uses: conda-incubator/setup-miniconda@v3
+        with:
+          miniforge-version: latest
+
+      - name: Install and activate locked conda environment
+        run: |
+          conda install conda-lock
+          conda-lock install --name test ${MODULE_PATH}/conda-lock.yml
+
+      # Update this step as needed to download the desired data
+      - name: Download test data
+        run: ./download-data.py --test-data --format AnnData
+
+      - name: Run analysis module
+        run: |
+          cd ${MODULE_PATH}
+          # run module script(s) here

analyses/celltype-glioblastoma/.dockerignore

@@ -0,0 +1,8 @@
+# Ignore everything by default
+*
+
+# Include specific files in the docker environment
+!/renv.lock
+!/requirements.txt
+!/environment.yml
+!/conda-lock.yml

analyses/celltype-glioblastoma/.gitignore

@@ -0,0 +1,7 @@
+# Results should not be committed
+/results/*
+!/results/README.md
+
+# Ignore the scratch directory (but keep it present)
+/scratch/*
+!/scratch/.gitkeep

analyses/celltype-glioblastoma/Dockerfile

@@ -0,0 +1,10 @@
+# A template docker file for creating a new analysis
+FROM ubuntu:22.04
+
+# Labels following the Open Containers Initiative (OCI) recommendations
+# For more information, see https://specs.opencontainers.org/image-spec/annotations/?v=v1.0.1
+LABEL org.opencontainers.image.authors="OpenScPCA scpca@ccdatalab.org"
+LABEL org.opencontainers.image.source="https://github.com/AlexsLemonade/OpenScPCA-analysis/tree/main/templates/analysis-module"
+
+# Set an environment variable to allow checking if we are in an OpenScPCA container
+ENV OPENSCPCA_DOCKER=TRUE

analyses/celltype-glioblastoma/README.md

@@ -0,0 +1,38 @@
+# Glioblastoma cell type identification
+
+## Description
+
+This module aims to identify the cell types present and provide labels for 16 pediatric Glioblastoma samples. This is split into the following objectives:
+- [ ] Visualise the data quality of each sample using standard quality control metrics (gene count, mitochondrial reads etc.)
+- [ ] Create a linear regression model using CellTypist with Glioblastoma atlas 'GBMap' (Core) as reference data
+- [ ] Apply the linear regression model using CellTypist to each sample individually
+- [ ] Use the pre-generated UMAP to cluster the cells and isolate clusters with low model confidence
+- [ ] Manually label the low-confidence clusters using known marker genes and differential expression analysis
+- [ ] Provide final cell labels for each sample 
+
+## Usage
+
+In progress
+
+## Input files
+
+The input is dependent on the output fles run from `download-data.py`. 
+
+```
+./download-data.py --projects SCPCP00001 --format anndata
+```
+
+This downloads the `Anndata` files for Glioblastoma samples.
+
+
+## Output files
+
+In progress
+
+## Software requirements
+
+The analysis will be done in Python version 3.11 primarily using the `scanpy' and `CellTypist' packages. 
+
+## Computational resources
+
+TBC

analyses/celltype-glioblastoma/environment.yml

@@ -0,0 +1,17 @@
+name: "openscpca-celltype-glioblastoma"
+channels:
+  - conda-forge
+  - bioconda
+platforms:
+  - linux-64
+  - osx-64
+  - osx-arm64
+dependencies:
+  - python=3.11
+  - awscli>=2.15
+  - conda-lock>=2.5
+  - jupyter=1.0
+  - jq>=1.7
+  - pre-commit>=3.6
+  - session-info>=1.0
+  - celltypist