Overview | When to Use methscope-cli | Build methscope-cli | Example Commands | Cell-Type Prediction | Generate a Feature Matrix | Deconvolution | Upscaling | Runnable example | Relationship to the R Package | Notes on Input Order

Overview | Install command-line dependencies | Choose the correct CpG reference | Convert M/U count BED files | Convert ALLC files | Convert beta or methylation fraction files | Convert binary BED tracks | Use the converted file in MethScope | Troubleshooting | Related resources

Generate an MRMP reference | Prerequisites | Step 1: Binarize the reference .cg file | Step 2: Build MRMP pattern definitions | Step 3: Pack patterns into a .cm reference | Build MRMP references from cell-type pseudobulks | Step 1: Create pseudobulk profiles per cell type | Step 2: Repeat per cell type and build MRMPs

Overview | 1) Generate cell/pixel/sample MRMP embeddings | 2) Perform cell-type annotation | 3) Train a custom classification model | 4) Visualize prediction results | 5) Cell-type deconvolution | 6) Unsupervised clustering (Seurat example) | 7) Upscaling Network (10k-CpG block example) | 7.1 Prepare model input (x) and target (y) | 7.2 Train decoder (Python) | 7.3 Evaluate | Related resources

Overview | Using a pretrained bundle | Bundling and unbundling | Bundle format (technical) | Models in R

Overview | Install for Codex | Example Codex prompts | Install for Claude-style project context | What the skill knows | Update the skill