How ablab/metagt works
Project type and purpose
• Project Type: A bioinformatics workflow/pipeline implemented in Nextflow (DSL2).
• Purpose: “MetaGT” performs assembly and quantification of metatranscriptome data by leveraging metagenome assemblies. It orchestrates a series of analysis steps (QC, assembly, annotation, alignment, clustering, quantification) using containerized processes.
File structure
• Top‐level files:
• nextflow.config, main.nf, nextflow_schema.json → Nextflow pipeline entry point and configuration
• environment.yml → Conda environment spec
• bin/ → standalone Python scripts for small utility tasks (name changes, extraction, parsing)
• conf/ → pipeline configuration profiles (base, test, test_full)
• modules/ → DSL2 modules organized under “local” (custom processes) and “nf-core” (community-maintained software wrappers)
• subworkflows/ → DSL2 subworkflows for assembling genome/transcriptome
• assets/ → static templates and configs (email templates, MultiQC config)
• lib/ → Groovy support libraries (schema definitions, header injection)
• data/ → example/reference data files
• Patterns & architectural choices:
• DSL2 modularization—“modules” and “subworkflows” directories
• One container per process (Docker/Singularity managed by Nextflow)
• Profiles for different environments in conf/
• Reusable utility scripts in bin/
README for additional insights
• Describes pipeline steps, software dependencies (Docker, Conda, Nextflow), and how modules map to tools (FastQC, metaSPAdes/rnaSPAdes, Prokka, minimap2, TransDecoder, MMseqs2, Kallisto).
• Emphasizes reproducibility and portability via containers and Conda.