# Complete CLI reference

[User guide](index.md) · [Workflow guide](workflow.md)

Generated by `python scripts/generate_cli_docs.py`. Run `metapathways COMMAND --help` for the help of your installed revision.

## prepare_test

```text
usage: metapathways [-h] -o OUTPUT_DIR

Copy the bundled test inputs and reference seeds into a writable workspace.

options:
  -h, --help            show this help message and exit
  -o OUTPUT_DIR, --output_dir OUTPUT_DIR
                        workspace for cami-test/ inputs and MPDB/ reference seeds
```

## run

```text
usage: Metapathways run [options]

Minimum REQUIRED Command:
MetaPathways run -i INPUT_FILE -o OUTPUT_DIR -d REFDB_DIR

options:
  -h, --help            show this help message and exit
  --dryrun              show the execution plan without running tasks

Minimum Required Arguments:
  -i INPUT_FILE, --input_file INPUT_FILE
                        path to the input fasta file/input dir [REQUIRED]
  -o OUTPUT_DIR, --output_dir OUTPUT_DIR
                        path to the output directory [REQUIRED]
  -d REFDB_DIR, --refdb_dir REFDB_DIR
                        path to the reference DB [REQUIRED]

Quality Controls Arguments:
  --input_format {fasta,fasta-amino}
                        Input format, FASTA support only [fasta]
  --qc_min_length QC_MIN_LENGTH
                        Minimum length for quality control [180]
  --qc_delete_replicates {yes,no}
                        Delete replicates in quality control [yes]

ORF Prediction Arguments:
  --orf_strand {pos,neg,both}
                        Strand for ORF prediction [both]
  --orf_algorithm ORF_ALGORITHM
                        Algorithm for ORF prediction, Prodigal support only [prodigal]
  --orf_min_length ORF_MIN_LENGTH
                        Minimum ORF length [60]
  --orf_translation_table ORF_TRANSLATION_TABLE
                        Translation table for ORF prediction, see Prodigal for translation tables [11]
  --orf_mode {single,meta}
                        Mode for ORF prediction [meta]

Functional Annotation Arguments:
  --annotation_algorithm {FAST,BLAST}
                        Algorithm for ORF annotation [FAST]
  --annotation_dbs ANNOTATION_DBS [ANNOTATION_DBS ...]
                        Database(s) for annotation, space-separated list [swissprot]
  --annotation_min_bsr ANNOTATION_MIN_BSR
                        Minimum BSR for annotation [0.4]
  --annotation_max_evalue ANNOTATION_MAX_EVALUE
                        Maximum e-value for annotation [0.000001]
  --annotation_min_score ANNOTATION_MIN_SCORE
                        Minimum score for annotation [20]
  --annotation_min_length ANNOTATION_MIN_LENGTH
                        Minimum length for annotation [45]
  --annotation_max_hits ANNOTATION_MAX_HITS
                        Maximum hits for annotation [5]
  --annotation_run_mode {default,pervol}
                        Run mode for annotation, FAST only [pervol]

rRNA Annotation Arguments:
  --rRNA_refdbs RRNA_REFDBS [RRNA_REFDBS ...]
                        Reference databases for rRNA annotation, space-separated list
                        [SILVA_138.1_LSURef_NR99_tax_silva_trunc SILVA_138.1_SSURef_NR99_tax_silva_trunc]
  --rRNA_max_evalue RRNA_MAX_EVALUE
                        Maximum e-value for rRNA annotation [0.000001]
  --rRNA_min_identity RRNA_MIN_IDENTITY
                        Minimum identity for rRNA annotation [20]
  --rRNA_min_bitscore RRNA_MIN_BITSCORE
                        Minimum bitscore for rRNA annotation [50]

Read Mapping Arguments (single sample support only):
  -1 FWD_FASTQ, --fastq FWD_FASTQ
                        location of the raw fastq file, either forward or interleaved
  -2 REV_FASTQ, --rev_fastq REV_FASTQ
                        location of the raw reverse fastq file, if separate paired-end
  --interleaved         if paired-end is interleaved [False]

Pipeline Step Arguments:
  --PREPROCESS_INPUT {yes,skip,redo}
                        Step: PREPROCESS_INPUT [yes]
  --ORF_PREDICTION {yes,skip,redo}
                        Step: ORF_PREDICTION [yes]
  --FILTER_AMINOS {yes,skip,redo}
                        Step: FILTER_AMINOS [yes]
  --SCAN_rRNA {yes,skip,redo}
                        Step: SCAN_rRNA [yes]
  --SCAN_tRNA {yes,skip,redo}
                        Step: SCAN_tRNA [yes]
  --FUNC_SEARCH {yes,skip,redo}
                        Step: FUNC_SEARCH [yes]
  --PARSE_FUNC_SEARCH {yes,skip,redo}
                        Step: PARSE_FUNC_SEARCH [yes]
  --ANNOTATE_ORFS {yes,skip,redo}
                        Step: ANNOTATE_ORFS [yes]
  --GENBANK_FILE {yes,skip,redo}
                        Step: GENBANK_FILE [yes]
  --CREATE_ANNOT_REPORTS {yes,skip,redo}
                        Step: CREATE_ANNOT_REPORTS [yes]
  --PATHOLOGIC_INPUT {yes,skip,redo}
                        Step: PATHOLOGIC_INPUT [yes]
  --COMPUTE_TPM {yes,skip,redo}
                        Step: COMPUTE_TPM [yes]
  --force_redo          Redo all steps [False]

Execution Resources:
  --max_cpus MAX_CPUS   total CPU budget [local: available CPUs; Slurm: no aggregate cap]
  --memory MEMORY       memory reservation per task [16 GB]
  --max_memory MAX_MEMORY
                        total memory budget [local: available memory; Slurm: no aggregate cap]
  --max_tasks MAX_TASKS
                        maximum submitted tasks, including queued/running [local: CPU budget; Slurm: 4]
  --executor {local,slurm}
                        execution backend [local]; Slurm uses your logged-in cluster identity
  --account ACCOUNT     Slurm allocation/account
  --partition PARTITION
                        Slurm partition [cluster default]
  --qos QOS             Slurm quality of service
  --reservation RESERVATION
                        Slurm reservation
  --time_limit TIME_LIMIT
                        Slurm walltime per task [24h]
  --submit_rate SUBMIT_RATE
                        maximum Slurm submissions per minute [6]
  --work_dir WORK_DIR   custom Nextflow work directory; retained after the run
  --conda_cache CONDA_CACHE
                        custom Conda cache directory; retained after the run
  --keep_work           retain automatically allocated work/cache directories

Miscellaneous Arguments:
  -s SAMPLES [SAMPLES ...], --samples SAMPLES [SAMPLES ...]
                        process only specific samples, space-separated list
  -t THREADS, --threads THREADS
                        threads per capable tool [8], capped by --max_cpus; serial stages use one CPU
  -v, --verbose         print more information on the stdout
  --test                use test values for all arguments
```

## analysis_wf

```text
usage: Metapathways analysis_wf [options]

Annotate multiple metagenomes, split MAGs, build PGDBs and generate a combined report.

options:
  -h, --help            show this help message and exit
  --dryrun              show the execution plan without running tasks
  --manifest MANIFEST   TSV with sample_id, assembly, read_layout, reads_1, reads_2, mag_map
  --reads_dir READS_DIR
                        Flat reads directory; -i then names the assemblies directory
  --mag_maps_dir MAG_MAPS_DIR
                        Flat contig-to-MAG map directory; -i then names the assemblies directory
  --no_reads            Explicitly omit read mapping during automatic discovery
  --no_mags             Explicitly omit MAG splitting during automatic discovery
  --skip_ptools         Omit community and MAG PGDB construction
  --compact_results     Use task scratch, archive PGDBs/diagnostics, and remove completed sample intermediates
  --scratch_dir SCRATCH_DIR
                        Worker-local scratch directory for compact mode [Slurm: SLURM_TMPDIR; local: system temporary directory]
  --image IMAGE         Pathway Tools SIF [registered by build_pt]
  --taxon_id TAXON_ID   Override PGDB NCBI taxon in private inputs; applies to every selected entity
  --taxonomic_scope {all,bacteria,archaea,eukaryotes}
                        Named PGDB taxon override; all means cellular life; euks aliases eukaryotes. Taxonomic pruning is enabled by default. Default: all (cellular life).
  --no_transport_inference
                        Disable TIP transport inference
  --taxprune            Enable taxonomic pruning [default]
  --no_taxprune         Disable taxonomic pruning and perform unpruned rescoring
  --ptools_memory PTOOLS_MEMORY
                        Optional PGDB memory override [same as --memory]

Minimum Required Arguments:
  -i INPUT_FILE, --input_file INPUT_FILE
                        dataset root, or assemblies directory with --reads_dir/--mag_maps_dir; alternative: --manifest
  -o OUTPUT_DIR, --output_dir OUTPUT_DIR
                        path to the output directory [REQUIRED]
  -d REFDB_DIR, --refdb_dir REFDB_DIR
                        path to the reference DB [REQUIRED]

Quality Controls Arguments:
  --input_format {fasta,fasta-amino}
                        Input format, FASTA support only [fasta]
  --qc_min_length QC_MIN_LENGTH
                        Minimum length for quality control [180]
  --qc_delete_replicates {yes,no}
                        Delete replicates in quality control [yes]

ORF Prediction Arguments:
  --orf_strand {pos,neg,both}
                        Strand for ORF prediction [both]
  --orf_algorithm ORF_ALGORITHM
                        Algorithm for ORF prediction, Prodigal support only [prodigal]
  --orf_min_length ORF_MIN_LENGTH
                        Minimum ORF length [60]
  --orf_translation_table ORF_TRANSLATION_TABLE
                        Translation table for ORF prediction, see Prodigal for translation tables [11]
  --orf_mode {single,meta}
                        Mode for ORF prediction [meta]

Functional Annotation Arguments:
  --annotation_algorithm {FAST,BLAST}
                        Algorithm for ORF annotation [FAST]
  --annotation_dbs ANNOTATION_DBS [ANNOTATION_DBS ...]
                        Database(s) for annotation, space-separated list [swissprot]
  --annotation_min_bsr ANNOTATION_MIN_BSR
                        Minimum BSR for annotation [0.4]
  --annotation_max_evalue ANNOTATION_MAX_EVALUE
                        Maximum e-value for annotation [0.000001]
  --annotation_min_score ANNOTATION_MIN_SCORE
                        Minimum score for annotation [20]
  --annotation_min_length ANNOTATION_MIN_LENGTH
                        Minimum length for annotation [45]
  --annotation_max_hits ANNOTATION_MAX_HITS
                        Maximum hits for annotation [5]
  --annotation_run_mode {default,pervol}
                        Run mode for annotation, FAST only [pervol]

rRNA Annotation Arguments:
  --rRNA_refdbs RRNA_REFDBS [RRNA_REFDBS ...]
                        Reference databases for rRNA annotation, space-separated list
                        [SILVA_138.1_LSURef_NR99_tax_silva_trunc SILVA_138.1_SSURef_NR99_tax_silva_trunc]
  --rRNA_max_evalue RRNA_MAX_EVALUE
                        Maximum e-value for rRNA annotation [0.000001]
  --rRNA_min_identity RRNA_MIN_IDENTITY
                        Minimum identity for rRNA annotation [20]
  --rRNA_min_bitscore RRNA_MIN_BITSCORE
                        Minimum bitscore for rRNA annotation [50]

Pipeline Step Arguments:
  --PREPROCESS_INPUT {yes,skip,redo}
                        Step: PREPROCESS_INPUT [yes]
  --ORF_PREDICTION {yes,skip,redo}
                        Step: ORF_PREDICTION [yes]
  --FILTER_AMINOS {yes,skip,redo}
                        Step: FILTER_AMINOS [yes]
  --SCAN_rRNA {yes,skip,redo}
                        Step: SCAN_rRNA [yes]
  --SCAN_tRNA {yes,skip,redo}
                        Step: SCAN_tRNA [yes]
  --FUNC_SEARCH {yes,skip,redo}
                        Step: FUNC_SEARCH [yes]
  --PARSE_FUNC_SEARCH {yes,skip,redo}
                        Step: PARSE_FUNC_SEARCH [yes]
  --ANNOTATE_ORFS {yes,skip,redo}
                        Step: ANNOTATE_ORFS [yes]
  --GENBANK_FILE {yes,skip,redo}
                        Step: GENBANK_FILE [yes]
  --CREATE_ANNOT_REPORTS {yes,skip,redo}
                        Step: CREATE_ANNOT_REPORTS [yes]
  --PATHOLOGIC_INPUT {yes,skip,redo}
                        Step: PATHOLOGIC_INPUT [yes]
  --COMPUTE_TPM {yes,skip,redo}
                        Step: COMPUTE_TPM [yes]
  --force_redo          Redo all steps [False]

Execution Resources:
  --max_cpus MAX_CPUS   total CPU budget [local: available CPUs; Slurm: no aggregate cap]
  --memory MEMORY       memory reservation per task [16 GB]
  --max_memory MAX_MEMORY
                        total memory budget [local: available memory; Slurm: no aggregate cap]
  --max_tasks MAX_TASKS
                        maximum submitted tasks, including queued/running [local: CPU budget; Slurm: 4]
  --executor {local,slurm}
                        execution backend [local]; Slurm uses your logged-in cluster identity
  --account ACCOUNT     Slurm allocation/account
  --partition PARTITION
                        Slurm partition [cluster default]
  --qos QOS             Slurm quality of service
  --reservation RESERVATION
                        Slurm reservation
  --time_limit TIME_LIMIT
                        Slurm walltime per task [24h]
  --submit_rate SUBMIT_RATE
                        maximum Slurm submissions per minute [6]
  --work_dir WORK_DIR   custom Nextflow work directory; retained after the run
  --conda_cache CONDA_CACHE
                        custom Conda cache directory; retained after the run
  --keep_work           retain automatically allocated work/cache directories

Miscellaneous Arguments:
  -t THREADS, --threads THREADS
                        threads per capable tool [8], capped by --max_cpus; serial stages use one CPU
  -v, --verbose         print more information on the stdout
```

## build_db

```text
usage: metapathways [-h] [-d PATH] [--func [CATEGORICAL ...]] [-a ALIGNER] [--skip_pt_screen]
                    [--screen_image SCREEN_IMAGE] [--metacyc_source METACYC_SOURCE] [-t INT]
                    [--dryrun] [--snakemake [SNAKEMAKE ...]] [--test] [--max_cpus MAX_CPUS]
                    [--memory MEMORY] [--max_memory MAX_MEMORY] [--max_tasks MAX_TASKS]
                    [--executor {local,slurm}] [--account ACCOUNT] [--partition PARTITION]
                    [--qos QOS] [--reservation RESERVATION] [--time_limit TIME_LIMIT]
                    [--submit_rate SUBMIT_RATE] [--work_dir WORK_DIR] [--conda_cache CONDA_CACHE]
                    [--keep_work]

automated database install

options:
  -h, --help            show this help message and exit
  -t INT, --threads INT
                        total database-build CPU budget [available CPUs]
  --dryrun              show the database execution plan without running tasks
  --snakemake [SNAKEMAKE ...]
                        legacy compatibility flags; use the resource flags for new runs
  --test                build test SwissProt/SILVA references; use -d for the prepared MPDB
                        directory

database arguments:
  -d PATH, --refdb_dir PATH
                        path to save the reference DB, [DEFAULT "./"]
  --func [CATEGORICAL ...]
                        functional references, select any combination from ['swissprot', 'cazy',
                        'eggnog', 'uniref50', 'uniref90', 'metacyc'], [DEFAULT ['swissprot']]
  -a ALIGNER, --aligner ALIGNER
                        local aligner to index for, select one of ['fast', 'blast'], [DEFAULT
                        fast]
  --skip_pt_screen      Skip default PTools reaction compatibility screening for MetaCyc
  --screen_image SCREEN_IMAGE
                        PTools SIF for screening a MetaCyc directory [registered SIF]
  --metacyc_source METACYC_SOURCE
                        licensed MetaCyc data directory or Pathway Tools SIF [registered SIF when
                        --func includes metacyc]

Execution Resources:
  --max_cpus MAX_CPUS   total CPU budget [local: available CPUs; Slurm: no aggregate cap]
  --memory MEMORY       memory reservation per task [16 GB]
  --max_memory MAX_MEMORY
                        total memory budget [local: available memory; Slurm: no aggregate cap]
  --max_tasks MAX_TASKS
                        maximum submitted tasks, including queued/running [local: CPU budget;
                        Slurm: 4]
  --executor {local,slurm}
                        execution backend [local]; Slurm uses your logged-in cluster identity
  --account ACCOUNT     Slurm allocation/account
  --partition PARTITION
                        Slurm partition [cluster default]
  --qos QOS             Slurm quality of service
  --reservation RESERVATION
                        Slurm reservation
  --time_limit TIME_LIMIT
                        Slurm walltime per task [24h]
  --submit_rate SUBMIT_RATE
                        maximum Slurm submissions per minute [6]
  --work_dir WORK_DIR   custom Nextflow work directory; retained after the run
  --conda_cache CONDA_CACHE
                        custom Conda cache directory; retained after the run
  --keep_work           retain automatically allocated work/cache directories
```

## mag_split

```text
usage: Metapathways mag_split [options]

Minimum REQUIRED Command:
Metapathways mag_split -o output_dir -m contig_mag_map

options:
  -h, --help            show this help message and exit
  -o OUTPUT_DIR, --output_dir OUTPUT_DIR
                        path where MP output was saved [REQUIRED]
  -m MAG_MAP, --contig_mag_map MAG_MAP
                        TSV file that contains contig-to-MAG mapping [REQUIRED]

Execution Resources:
  --max_cpus MAX_CPUS   total CPU budget [local: available CPUs; Slurm: no aggregate cap]
  --memory MEMORY       memory reservation per task [16 GB]
  --max_memory MAX_MEMORY
                        total memory budget [local: available memory; Slurm: no aggregate cap]
  --max_tasks MAX_TASKS
                        maximum submitted tasks, including queued/running [local: CPU budget; Slurm: 4]
  --executor {local,slurm}
                        execution backend [local]; Slurm uses your logged-in cluster identity
  --account ACCOUNT     Slurm allocation/account
  --partition PARTITION
                        Slurm partition [cluster default]
  --qos QOS             Slurm quality of service
  --reservation RESERVATION
                        Slurm reservation
  --time_limit TIME_LIMIT
                        Slurm walltime per task [24h]
  --submit_rate SUBMIT_RATE
                        maximum Slurm submissions per minute [6]
  --work_dir WORK_DIR   custom Nextflow work directory; retained after the run
  --conda_cache CONDA_CACHE
                        custom Conda cache directory; retained after the run
  --keep_work           retain automatically allocated work/cache directories
```

## build_pt

```text
usage: metapathways build_pt [-h] -i INSTALLER [--ptools_version PTOOLS_VERSION] [-d REFDB_DIR]
                             [--skip_pt_screen] [-a {fast,blast}] [-o OUTPUT_DIR] [-t THREADS]
                             [--dryrun] [--max_cpus MAX_CPUS] [--memory MEMORY]
                             [--max_memory MAX_MEMORY] [--max_tasks MAX_TASKS]
                             [--executor {local,slurm}] [--account ACCOUNT]
                             [--partition PARTITION] [--qos QOS] [--reservation RESERVATION]
                             [--time_limit TIME_LIMIT] [--submit_rate SUBMIT_RATE]
                             [--work_dir WORK_DIR] [--conda_cache CONDA_CACHE] [--keep_work]

Build a Pathway Tools SIF from a local Linux installer using Nextflow and Apptainer.

options:
  -h, --help            show this help message and exit
  -i INSTALLER, --installer INSTALLER
                        local Pathway Tools Linux x86-64 installer
  --ptools_version PTOOLS_VERSION
                        release number if the installer was renamed; otherwise inferred from its
                        filename
  -d REFDB_DIR, --refdb_dir REFDB_DIR
                        also export and prepare the licensed MetaCyc reference in this MPDB after
                        building the SIF
  --skip_pt_screen      Skip default reaction compatibility screening when building MetaCyc with
                        -d
  -a {fast,blast}, --aligner {fast,blast}
                        MetaCyc reference index format with -d [fast]
  -o OUTPUT_DIR, --output_dir OUTPUT_DIR
                        container directory [~/.local/share/metapathways/containers]
  -t THREADS, --threads THREADS
                        CPUs for image compression [2]; Pathway Tools uses one CPU
  --dryrun              write and show the build plan without building or registering

Execution Resources:
  --max_cpus MAX_CPUS   total CPU budget [local: available CPUs; Slurm: no aggregate cap]
  --memory MEMORY       memory reservation per task [4 GB]
  --max_memory MAX_MEMORY
                        total memory budget [local: available memory; Slurm: no aggregate cap]
  --max_tasks MAX_TASKS
                        maximum submitted tasks, including queued/running [local: CPU budget;
                        Slurm: 4]
  --executor {local,slurm}
                        execution backend [local]; Slurm uses your logged-in cluster identity
  --account ACCOUNT     Slurm allocation/account
  --partition PARTITION
                        Slurm partition [cluster default]
  --qos QOS             Slurm quality of service
  --reservation RESERVATION
                        Slurm reservation
  --time_limit TIME_LIMIT
                        Slurm walltime per task [24h]
  --submit_rate SUBMIT_RATE
                        maximum Slurm submissions per minute [6]
  --work_dir WORK_DIR   custom Nextflow work directory; retained after the run
  --conda_cache CONDA_CACHE
                        custom Conda cache directory; retained after the run
  --keep_work           retain automatically allocated work/cache directories
```

## screen_pt

```text
usage: metapathways [-h] -d REFDB_DIR -o OUTPUT_DIR [--image IMAGE] [--publish]
                    [--reactions REACTIONS [REACTIONS ...]] [--batch_size BATCH_SIZE]
                    [--max_tasks MAX_TASKS] [--confirm_runs CONFIRM_RUNS] [--timeout TIMEOUT]
                    [--scratch_dir SCRATCH_DIR]

Screen explicit MPDB reaction assignments against a licensed PTools SIF

options:
  -h, --help            show this help message and exit
  -d REFDB_DIR, --refdb_dir REFDB_DIR
  -o OUTPUT_DIR, --output_dir OUTPUT_DIR
  --image IMAGE         PTools SIF [registered build_pt image]
  --publish             Save a completed full screen compatibility list in the MPDB
  --reactions REACTIONS [REACTIONS ...]
                        Optional reaction IDs for a targeted screen
  --batch_size BATCH_SIZE
  --max_tasks MAX_TASKS
                        Concurrent isolated containers [1]
  --confirm_runs CONFIRM_RUNS
  --timeout TIMEOUT     Seconds per build [1800]; timeout is inconclusive
  --scratch_dir SCRATCH_DIR
                        Local temporary storage for private PGDB builds
```

## ptools

```text
usage: Metapathways ptools [options]

Minimum REQUIRED Command:
Metapathways ptools -o output_dir

options:
  -h, --help            show this help message and exit
  -o OUTPUT_DIR, --output_dir OUTPUT_DIR
                        path where MP output was saved [REQUIRED]
  --tag TAG             Custom name for ePGDB [optional]
  --container           Flag only used in containerized env [special flag]
  --taxprune            Enable taxonomic pruning [default]
  --no_taxprune         Disable taxonomic pruning and perform unpruned rescoring
  --taxon_id TAXON_ID   Override PGDB NCBI taxon in private inputs; applies to every selected entity
  --taxonomic_scope {all,bacteria,archaea,eukaryotes}
                        Named PGDB taxon override; all means cellular life; euks aliases eukaryotes. Taxonomic pruning is enabled by default. Default: all (cellular life).
  --no_transport_inference
                        Disable TIP transport inference (SIF only)
  --entity ENTITY       Build only community or the specified MAG ID
  --image IMAGE         Pathway Tools SIF [registered by build_pt]

Execution Resources:
  --max_cpus MAX_CPUS   total CPU budget [local: available CPUs; Slurm: no aggregate cap]
  --memory MEMORY       memory reservation per task [4 GB]
  --max_memory MAX_MEMORY
                        total memory budget [local: available memory; Slurm: no aggregate cap]
  --max_tasks MAX_TASKS
                        maximum submitted tasks, including queued/running [local: CPU budget; Slurm: 4]
  --executor {local,slurm}
                        execution backend [local]; Slurm uses your logged-in cluster identity
  --account ACCOUNT     Slurm allocation/account
  --partition PARTITION
                        Slurm partition [cluster default]
  --qos QOS             Slurm quality of service
  --reservation RESERVATION
                        Slurm reservation
  --time_limit TIME_LIMIT
                        Slurm walltime per task [24h]
  --submit_rate SUBMIT_RATE
                        maximum Slurm submissions per minute [6]
  --work_dir WORK_DIR   custom Nextflow work directory; retained after the run
  --conda_cache CONDA_CACHE
                        custom Conda cache directory; retained after the run
  --keep_work           retain automatically allocated work/cache directories
```

## report

```text
usage: metapathways [-h] -o OUTPUT_DIR [--serve] [--no-rebuild] [--port PORT] [--no-browser]

Build a navigable report from existing MP outputs; no analyses are rerun.

options:
  -h, --help            show this help message and exit
  -o OUTPUT_DIR, --output_dir OUTPUT_DIR
                        sample output or parent containing sample outputs
  --serve               open the local searchable portal and serve until Ctrl-C
  --no-rebuild          use an existing report snapshot
  --port PORT           local port [automatically selected]
  --no-browser          print the local URL without opening a browser
```