How imallona/repeats works
Main Components Organization
a) Input Layer:
• Configuration files (workflow/configs/*.yaml)
• Reference data inputs (RepeatMasker and GENCODE annotations)
b) Workflow Layer (Core Processing):
Main Snakemake workflows as connected modules:
• Data Acquisition (data_acquisition.snmk)
• Reference Making (reference_making.snmk)
• Analysis Pipelines:
• Alevin Pipeline (alevin.snmk)
• STARsolo Pipeline (starsolo.snmk)
• Counting Pipeline (counting.snmk)
• Simulation Pipeline (simulations.snmk)
c) Processing Scripts Layer:
• Cell Barcode Processing Scripts
• Integration Scripts (R Markdown files)
• Summary Generation Scripts
d) Output Layer:
• Analysis Results
• Reports
• Integrated Data
Flow Directions
• Raw data → Data Acquisition → Processing Pipelines
• Reference Data → Reference Making → Processing Pipelines
• Processing Results → Integration Scripts → Final Reports
Key Features to Highlight
a) Conda Environment Integration:
• Connection to workflow/envs/repeats.yaml
• Conda-managed dependencies
b) Pipeline Variants:
• Different analysis paths for various sequencing technologies
• Parallel processing capabilities
c) Configuration Management:
• Configuration validation through workflow/schemas/config.schema.yaml
• How configs feed into different pipeline branches
Additional Details
• Version information (v0.1 vs. current)
• Integration points with Zenodo for data sharing
• Where different cell types are processed
• Parallel processing capabilities
• High-level workflow overview
• Detailed processing steps
• Technical implementation details
• Data flow patterns
Remember to:
• Clear separation between different analysis approaches
• Where configuration changes can affect the workflow
• Points where quality control and validation occur
This system design should clearly communicate how the pipeline processes different types of single-cell RNA-seq data while maintaining flexibility for different input types and analysis requirements.