MetaPathways
Getting started
What MetaPathways does
Installation and first run
1. Conda package with Mamba (preferred)
2. Quay: Docker or Apptainer
3. Local installation from GitHub
Try your own data
CAMI references
Getting started: from a terminal to your first result
What you will do
Before typing commands
Install on Linux x86-64
First installation check
Your first real assembly
Terms used in the guides
Quay containers: Docker and Apptainer
Docker three-sample test
Apptainer three-sample test
From the test to your own data
CAMI references
Test walkthrough: three tiny CAMI samples
1. Install MP and prepare the small reference database
Run all three samples first
2. Run the single-sample workflow
3. Run two distinct samples together
4. Check outputs and explore tables
5. Optional: include licensed Pathway Tools
Provenance and validation
Recorded three-sample validation
CAMI references
Tiny CAMI II test inputs
Source attribution
CAMI references
Citing CAMI and CAMI II
CAMI references
Reference-manager download
Analysis guides
Organize assemblies, reads and genome assignments
Decide what each sample means
Automatic discovery
Organize with links, without copying large reads
Custom manifest: keep every file where it is
Contig-to-genome map format
Prepare reference databases
Pathway Tools: licensing, image builds, databases and inference
Defaults and explicit choices
Understand the three different databases
Get the installer
Host prerequisites
Build and register once
Build the MPDB MetaCyc annotation reference at the same time
Run pathway inference
Choose a taxonomic scope
Explicit reaction blacklist
Transport inference and sequence-backed inputs
Outputs, warnings and failures
Screen reaction compatibility (maintainers)
Intermittent container startup failures
Complete multi-sample analysis
Analysis wf input layout
Custom analysis manifest
Outputs, restarting and exploration
Annotate your own data
One assembly
Read mapping
Multiple assemblies
Stage controls
MAGs and pathway inference
Build Pathway Tools once
MetaCyc from Pathway Tools
Community pathways
Add MAG pathways without reannotating
Command cookbook: run only the modules you need
Choose a command
prepare_test: try the included data
build_db: references before analysis
run: annotations and optional abundance
analysis_wf: one complete run for N samples
mag_split: reuse community annotation
build_pt and ptools: licensed setup, then inference
report: inspect and export without reanalysis
Resources, failure and restart
Resources and Slurm
Local execution is the default
Submit from a Slurm headnode
Logs, temporary files and restarting
Troubleshooting and reproducibility
Results and reference
Explore results, follow links and export tables
Open the report or start the explorer
View a remote report through SSH
Start with accounting
Follow one annotation through the results
Select useful subsets
Avoid accidental double counting
Share or archive a result
Reports and the EDA portal
Results schema and EDA portal
What the report does
Relationships
Tables and their row units
Ready-made explorer views
Subsetting examples
Missing, partial and historical outputs
Outputs that stay as source files
Exports and local service
Software architecture: local and HPC
Data flow: inputs to connected results
Detailed workflow and tool citations
How to cite your analysis
1. Prepare references and the optional Pathway Tools image
2. Plan and schedule work
3. Predict and annotate features
4. Measure abundance, build PGDBs and assemble reports
Tool and wrapper index
Reference databases
Supporting software and distribution
Full references
Workflow and output guide
Command boundaries
Annotation stages
Output layout
Dependency and cache behavior
Report limitations and error recovery
Complete CLI reference
prepare_test
run
analysis_wf
build_db
mag_split
build_pt
screen_pt
ptools
report
Benchmarking and supplementary statistics
Define the experiment before starting
Example full workflow
What is recorded automatically
Supplementary table structure
Runtime and resource figures
Save environment and machine information
Finish, inspect, then summarize
Compare one server with a Slurm cluster
Compact results on limited storage
CAMI references
Reproducibility and validation
Record the actual revision and inputs
Installation example versus biological validation
Historical benchmark provenance
Documentation and release records
CAMI references
Development
PR tester checklist
CAMI references
Releasing MetaPathways
Normal release using GitHub CI
Shared release controls (MetaPathways and SCARAB)
Enable Anaconda.org uploads
Local build and validation
Rebuilds and failed releases
Recover the interrupted v3.5.0 GitHub upload
Docker, Apptainer, and Quay
Security rebuilds without changing the application version
Feature-branch review before publication
Zenodo and software citation
Account-side release checklist
CAMI references
MetaPathways 3.5.2 release readiness
Registry and container status
Completed checks
Before merge and public release
CAMI references
Maintaining the documentation
Build and review locally
Connect Read the Docs
Shared diagram scale
MetaPathways
Index
Index