Continue getting closer.

.github/workflows/run-tests.yaml

@@ -0,0 +1,48 @@
+on:
+  push:
+    branches:
+      - master
+  pull_request:
+
+name: Build the tool
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+
+    steps:
+    - name: Checkout repo
+      uses: actions/checkout@v4
+
+    - name: Update NPM packages
+      run: npm i --include=dev
+
+    - name: Cache installed
+      id: installed
+      uses: actions/cache@v4
+      with:
+        path: extern/installed
+        key: deps-${{ hashFiles('extern/**/build.sh') }}
+
+    - name: Build igraph
+      if: steps.installed.outputs.cache-hit != 'true'
+      run: |
+        cd extern/igraph
+        bash build.sh
+
+    - name: Build the tool
+      run: |
+        cmake \
+            -S . \
+            -B build \
+            -DCMAKE_BUILD_TYPE=Release \
+            -DCMAKE_INSTALL_PREFIX=extern/igraph
+        cmake --build build
+
+    - name: Add knncolle tagfile
+      uses: wei/wget@v1
+      with:
+        args: -O matrix.mtx.gz https://github.com/kanaverse/random-test-files/releases/download/zeisel-brain-v1.0.0/matrix.mtx.gz
+
+    - name: Run the tool
+      run: ./build/scran --mito-list 19972-20005 matrix.mtx.gz

README.md

@@ -1,5 +1,7 @@
 # CLI for single-cell analyses
 
+![Unit tests](https://github.com/libscran/scran-cli/actions/workflows/run-tests.yaml/badge.svg)
+
 ## Overview
 
 This repository implements a command-line tool for single-cell RNA-seq data analysis from a Matrix Market file.
@@ -33,30 +35,31 @@ This will produce a `scran` executable in `build`, which can be run on the comma
 ```bash
 ./build/scran --help
 ## Single-cell RNA-seq analyses on the command-line
-## Usage: ./build/scran [OPTIONS] path
+## Usage: ./build/scran [OPTIONS] counts
 ## 
 ## Positionals:
-##   path TEXT REQUIRED          Path to the Matrix Market file
+##   counts TEXT REQUIRED        Path to the MatrixMarket file containing the counts.
 ## 
 ## Options:
 ##   -h,--help                   Print this help message and exit
-##   -t,--nthreads FLOAT=1       Number of threads to use (+2 for UMAP and t-SNE, which use their own threads)
-##   -o,--output TEXT=output     Path to the output directory
-##   --skip-output BOOLEAN=0     Run the analysis but do not save results
-##   --qc-nmads FLOAT=3          Number of MADs to use for filtering
-##   --hvg-span FLOAT=0.4        LOWESS span for variance modelling
-##   --hvg-num INT=2500          Number of HVGs to use for PCA
-##   --pca-num INT=25            Number of PCs to keep
-##   --nn-approx BOOLEAN=1       Whether to use an approximate neighbor search
-##   --snn-neighbors INT=10      Number of neighbors to use for the SNN graph
-##   --snn-scheme ENUM:value in {jaccard->2,number->1,ranked->0} OR {2,1,0}=0
-##                               Edge weighting scheme: ranked, number or jaccard
-##   --snn-res FLOAT=1           Resolution to use in multi-level community detection
-##   --tsne-perplexity FLOAT=30  Perplexity to use in t-SNE
-##   --tsne-iter INT=500         Number of iterations to use in t-SNE
-##   --umap-neighbors INT=15     Number of neighbors to use in the UMAP
-##   --umap-mindist FLOAT=0.01   Minimum distance to use in the UMAP
-##   --umap-epochs INT=500       Number of epochs to use in the UMAP
+##   -o,--output TEXT [output]   Path to the output directory. If empty, results are not saved.
+##   --mito-list TEXT            Comma-separated list of the 0-based row indices of the mitochondrial genes. Closed intervals are also accepted as 'X-Y'.
+##   --num-mads FLOAT [3]        Number of MADs to use for defining QC thresholds.
+##   --fit-span FLOAT [0.3]      LOWESS span for fitting the mean-variance trend.
+##   --num-hvgs INT [2500]       Number of HVGs to use for PCA.
+##   --num-pcs INT [25]          Number of PCs to keep.
+##   --nn-approx INT [1]         Whether to use an approximate neighbor search.
+##   --snn-neighbors INT [10]    Number of neighbors to use for the SNN graph.
+##   --snn-scheme TEXT:{ranked,number,jaccard} [ranked]
+##                               Edge weighting scheme for SNN graph construction.
+##   --snn-res FLOAT [0.5]       Resolution to use in multi-level community detection.
+##   --tsne-perplexity FLOAT [30]
+##                               Perplexity to use in t-SNE.
+##   --tsne-iter FLOAT [500]     Number of iterations to use in t-SNE.
+##   --umap-neighbors INT [15]   Number of neighbors to use in the UMAP.
+##   --umap-mindist FLOAT [0.1]  Minimum distance to use in the UMAP.
+##   --umap-epochs INT [500]     Number of epochs to use in the UMAP.
+##   -t,--nthreads INT [1]       Number of threads to use (+2 for UMAP and t-SNE, which use their own threads).
 ```
 
 ## Usage instructions

src/main.cpp

@@ -111,7 +111,7 @@ int main(int argc, char* argv []) {
         app.add_option("counts", all_opt.file_path, "Path to the MatrixMarket file containing the counts.")->required();
         app.add_option("-o,--output", all_opt.output, "Path to the output directory. If empty, results are not saved.")->default_val("output");
 
-        app.add_option("--mito-list", all_opt.mito_list, "Comma-separated list of the 0-based row indices of the mitochondrial genes")->default_val("");
+        app.add_option("--mito-list", all_opt.mito_list, "Comma-separated list of the 0-based row indices of the mitochondrial genes. Closed intervals are also accepted as 'X-Y'.")->default_val("");
         app.add_option("--num-mads", all_opt.num_mads, "Number of MADs to use for defining QC thresholds.")->default_val(3);
 
         app.add_option("--fit-span", all_opt.fit_span, "LOWESS span for fitting the mean-variance trend.")->default_val(0.3);