Command line interface

This part of the documentation describes the snakemake executable. Snakemake is primarily a command-line tool, so the snakemake executable is the primary way to execute, debug, and visualize workflows.

Important environment variables

Snakemake caches source files for performance and reproducibility. The location of this cache is determined by the appdirs package. If you want to change the location on a unix/linux system, you can define an override path via the environment variable XDG_CACHE_HOME.

Useful Command Line Arguments

If called with the number of cores to use, i.e.

$ snakemake --cores 1

Snakemake tries to execute the workflow specified in a file called Snakefile in the same directory (the Snakefile can be given via the parameter -s).

By issuing

$ snakemake -n

a dry-run can be performed. This is useful to test if the workflow is defined properly and to estimate the amount of needed computation.

Importantly, Snakemake can automatically determine which parts of the workflow can be run in parallel. By specifying more than one available core, i.e.

$ snakemake --cores 4

one can tell Snakemake to use up to 4 cores and solve a binary knapsack problem to optimize the scheduling of jobs. If the number is omitted (i.e., only --cores is given), the number of used cores is determined as the number of available CPU cores in the machine.

Snakemake workflows usually define the number of used threads of certain rules. Sometimes, it makes sense to overwrite the defaults given in the workflow definition. This can be done by using the --set-threads argument, e.g.,

$ snakemake --cores 4 --set-threads myrule=2

would overwrite whatever number of threads has been defined for the rule myrule and use 2 instead. Similarly, it is possible to overwrite other resource definitions in rules, via

$ snakemake --cores 4 --set-resources myrule:partition="foo"

Both mechanisms can be particularly handy when used in combination with non-local execution.

Non-local execution

Non-local execution on cluster or cloud infrastructure is implemented via plugins. The Snakemake plugin catalog lists available plugins and their documentation. In general, the configuration boils down to specifying an executor plugin (e.g. for SLURM or Kubernetes) and, if needed, a storage plugin (e.g. in order to use S3 for input and output files or in order to efficiently use a shared network filesystem). For maximizing the I/O performance over the network, it can be advisable to annotate the input file access patterns of rules. Snakemake provides lots of tunables for non-local execution, which can all be found under All Options and in the plugin descriptions of the Snakemake plugin catalog. In any case, the cluster or cloud specific configuration will entail lots of command line options to be chosen and set, which should be persisted in a profile.

Dealing with very large workflows

If your workflow has a lot of jobs, Snakemake might need some time to infer the dependencies (the job DAG) and which jobs are actually required to run. The major bottleneck involved is the filesystem, which has to be queried for existence and modification dates of files. To overcome this issue, Snakemake allows to run large workflows in batches. This way, fewer files have to be evaluated at once, and therefore the job DAG can be inferred faster. By running

$ snakemake --cores 4 --batch myrule=1/3

you instruct to only compute the first of three batches of the inputs of the rule myrule. To generate the second batch, run

$ snakemake --cores 4 --batch myrule=2/3

Finally, when running

$ snakemake --cores 4 --batch myrule=3/3

Snakemake will process beyond the rule myrule, because all of its input files have been generated, and complete the workflow. Obviously, a good choice of the rule to perform the batching is a rule that has a lot of input files and upstream jobs, for example a central aggregation step within your workflow. We advice all workflow developers to inform potential users of the best suited batching rule.

Profiles

Adapting Snakemake to a particular environment can entail many flags and options. Therefore, since Snakemake 4.1, it is possible to specify configuration profiles to be used to obtain default options. Since Snakemake 7.29, two kinds of profiles are supported:

• A global profile that is defined in a system-wide or user-specific configuration directory (on Linux, this will be $HOME/.config/snakemake and /etc/xdg/snakemake, you can find the answer for your system via snakemake --help).

• A workflow specific profile (introduced in Snakemake 7.29) that is defined via a flag (--workflow-profile) or searched in a default location (profiles/default) in the working directory or next to the Snakefile.

The workflow specific profile is meant to be used to define default options for a particular workflow, like providing constraints for certain custom resources the workflow uses (e.g. api_calls) or overwriting the threads and resource definitions of individual rules without modifying the workflow code itself. In contrast, the global profile is meant to be used to define default options for a particular environment, like the default cluster submission command or the default number of jobs to run in parallel.

For example, the command

$ snakemake --profile myprofile

would expect a folder myprofile in per-user and global configuration directories (on Linux, this will be $HOME/.config/snakemake and /etc/xdg/snakemake, you can find the answer for your system via snakemake --help). Alternatively, an absolute or relative path to the profile folder can be given. The default profile to use when no --profile argument is specified can also be set via the environment variable SNAKEMAKE_PROFILE, e.g. by specifying export SNAKEMAKE_PROFILE=myprofile in your ~/.bashrc or the system wide shell defaults means that the --profile flag can be omitted. In order unset the profile defined by this environment variable for individual runs without specifying and alternative profile you can provide the special value none, i.e. --profile none.

The profile folder is expected to contain a configuration file that file that defines default values for the Snakemake command line arguments. The file has to be named config.vX+.yaml with X denoting the minimum supported Snakemake major version (e.g. config.v8+.yaml). As fallback, it is also possible to provide a version agnostic config.yaml that matches any Snakemake version. For example, the file

executor: slurm
jobs: 100

would setup Snakemake to always submit to the SLURM cluster middleware and never use more than 100 parallel jobs in total. The profile can be used to set a default for each option of the Snakemake command line interface. For this, option --someoption becomes someoption: in the profile. The profile folder can additionally contain auxiliary files, e.g., jobscripts, or any kind of wrappers. See https://github.com/snakemake-profiles/doc for examples. If options accept multiple arguments these must be given as YAML list in the profile. If options expect structured arguments (like --default-resources RESOURCE=VALUE, --set-threads RULE=VALUE, or --set-resources RULE:RESOURCE=VALUE), those can be given as strings in the expected forms, i.e.

default-resources: mem_mb=200
set-threads: myrule=5
set-resources: myrule:mem=500MB

or as YAML maps, which is easier to read:

default-resources:
    mem_mb: 200
set-threads:
    myrule: 5
set-resources:
    myrule:
        mem: 500MB

All of these resource specifications can also be made dynamic, by using expressions and certain variables that are available. For details of the variables you can use, refer to the callable signatures given in the documentation sections on the specification of threads and dynamic resources. These enable config.yaml entries like:

default-resources:
    mem_mb: max(1.5 * input.size_mb, 100)
set-threads:
    myrule: max(input.size_mb / 5, 2)
set-resources:
    myrule:
        mem_mb: attempt * 200

Setting resources or threads via the profile is of course rather a job for the workflow profile instead of the global profile (as such settings are likely workflow specific).

Values in profiles can make use of globally available environment variables, e.g. the $USER variable. For example, the following would set the default prefix for storing local copies of remote storage files to a user specific directory

local-storage-prefix: /local/work/$USER/snakemake-scratch

Any such environment variables are automatically expanded when evaluating the profile.

Under https://github.com/snakemake-profiles/doc, you can find publicly available global profiles (e.g. for cluster systems). Feel free to contribute your own. Workflow specific profiles are either not shared at all, or can be distributed along with the workflow itself where it makes sense. For example, when the workflow has its Snakefile at workflow/Snakefile, the profile config should be placed at workflow/profiles/default/config.yaml.

Use templating in profiles

In Snakemake 7.30 or newer, when the profile starts with

__use_yte__: true

It will be treated as a YTE template and parsed accordingly. This can be handy to e.g. define values inside of the profile that are based on environment variables. For example, admins could use this to define user-specific settings. Another application would be the uniform redefinition of resource requirements for a larger set of rules in a workflow profile (see above). However, it should be noted that templated profiles are harder to keep free of errors and the profile author has to make sure that they always work correctly for the user.

Visualization

To visualize the workflow, one can use the option --dag. This creates a representation of the DAG in the graphviz dot language which has to be postprocessed by the graphviz tool dot. E.g. to visualize the DAG that would be executed, you can issue:

$ snakemake --dag | dot | display

For saving this to a file, you can specify the desired format:

$ snakemake --dag | dot -Tpdf > dag.pdf

To visualize the whole DAG regardless of the eventual presence of files, the forceall option can be used:

$ snakemake --forceall --dag | dot -Tpdf > dag.pdf

Of course the visual appearance can be modified by providing further command line arguments to dot.

Note: The DAG is printed in DOT format straight to the standard output, along with other print statements you may have in your Snakefile. Make sure to comment these other print statements so that dot can build a visual representation of your DAG.

All Options

All command line options can be printed by calling snakemake -h.

Snakemake is a Python based language and execution environment for GNU Make-like workflows.

usage: snakemake [-h] [--dry-run] [--profile PROFILE]
                 [--workflow-profile WORKFLOW_PROFILE] [--cache [RULE ...]]
                 [--snakefile FILE] [--cores N] [--jobs N] [--local-cores N]
                 [--resources NAME=INT [NAME=INT ...]]
                 [--set-threads RULE=THREADS [RULE=THREADS ...]]
                 [--max-threads MAX_THREADS]
                 [--set-resources RULE:RESOURCE=VALUE [RULE:RESOURCE=VALUE ...]]
                 [--set-scatter NAME=SCATTERITEMS [NAME=SCATTERITEMS ...]]
                 [--set-resource-scopes RESOURCE=[global|local]
                 [RESOURCE=[global|local] ...]]
                 [--default-resources [NAME=INT ...]]
                 [--preemptible-rules [PREEMPTIBLE_RULES ...]]
                 [--preemptible-retries PREEMPTIBLE_RETRIES]
                 [--configfile FILE [FILE ...]] [--config [KEY=VALUE ...]]
                 [--replace-workflow-config] [--envvars VARNAME [VARNAME ...]]
                 [--directory DIR] [--touch] [--keep-going]
                 [--rerun-triggers {code,input,mtime,params,software-env} [{code,input,mtime,params,software-env} ...]]
                 [--force] [--executor {local,dryrun,touch}] [--forceall]
                 [--forcerun [TARGET ...]]
                 [--consider-ancient RULE=INPUTITEMS [RULE=INPUTITEMS ...]]
                 [--prioritize TARGET [TARGET ...]]
                 [--batch RULE=BATCH/BATCHES] [--until TARGET [TARGET ...]]
                 [--omit-from TARGET [TARGET ...]] [--rerun-incomplete]
                 [--shadow-prefix DIR]
                 [--strict-dag-evaluation {cyclic-graph,functions,periodic-wildcards} [{cyclic-graph,functions,periodic-wildcards} ...]]
                 [--scheduler [{ilp,greedy}]] [--wms-monitor [WMS_MONITOR]]
                 [--wms-monitor-arg [NAME=VALUE ...]]
                 [--scheduler-ilp-solver {PULP_CBC_CMD}]
                 [--conda-base-path CONDA_BASE_PATH] [--no-subworkflows]
                 [--precommand PRECOMMAND] [--groups GROUPS [GROUPS ...]]
                 [--group-components GROUP_COMPONENTS [GROUP_COMPONENTS ...]]
                 [--report [FILE]] [--report-after-run]
                 [--report-stylesheet CSSFILE] [--reporter PLUGIN]
                 [--draft-notebook TARGET] [--edit-notebook TARGET]
                 [--notebook-listen IP:PORT] [--lint [{text,json}]]
                 [--generate-unit-tests [TESTPATH]] [--containerize]
                 [--export-cwl FILE] [--list-rules] [--list-target-rules]
                 [--dag [{dot,mermaid-js}]] [--rulegraph [{dot,mermaid-js}]]
                 [--filegraph] [--d3dag] [--summary] [--detailed-summary]
                 [--archive FILE] [--cleanup-metadata FILE [FILE ...]]
                 [--cleanup-shadow] [--skip-script-cleanup] [--unlock]
                 [--list-changes {params,code,input}] [--list-input-changes]
                 [--list-params-changes] [--list-untracked]
                 [--delete-all-output | --delete-temp-output]
                 [--keep-incomplete] [--drop-metadata] [--version]
                 [--printshellcmds] [--debug-dag] [--nocolor]
                 [--quiet [{all,host,progress,reason,rules} ...]]
                 [--print-compilation] [--verbose] [--force-use-threads]
                 [--allow-ambiguity] [--nolock] [--ignore-incomplete]
                 [--max-inventory-time SECONDS] [--trust-io-cache]
                 [--max-checksum-file-size SIZE] [--latency-wait SECONDS]
                 [--wait-for-free-local-storage WAIT_FOR_FREE_LOCAL_STORAGE]
                 [--wait-for-files [FILE ...]] [--wait-for-files-file FILE]
                 [--queue-input-wait-time SECONDS] [--notemp] [--all-temp]
                 [--unneeded-temp-files FILE [FILE ...]]
                 [--keep-storage-local-copies] [--not-retrieve-storage]
                 [--target-files-omit-workdir-adjustment]
                 [--allowed-rules ALLOWED_RULES [ALLOWED_RULES ...]]
                 [--max-jobs-per-timespan MAX_JOBS_PER_TIMESPAN]
                 [--max-jobs-per-second MAX_JOBS_PER_SECOND]
                 [--max-status-checks-per-second MAX_STATUS_CHECKS_PER_SECOND]
                 [--seconds-between-status-checks SECONDS_BETWEEN_STATUS_CHECKS]
                 [--retries RETRIES] [--wrapper-prefix WRAPPER_PREFIX]
                 [--default-storage-provider DEFAULT_STORAGE_PROVIDER]
                 [--default-storage-prefix DEFAULT_STORAGE_PREFIX]
                 [--local-storage-prefix LOCAL_STORAGE_PREFIX]
                 [--remote-job-local-storage-prefix REMOTE_JOB_LOCAL_STORAGE_PREFIX]
                 [--shared-fs-usage {input-output,persistence,software-deployment,source-cache,sources,storage-local-copies,none} [{input-output,persistence,software-deployment,source-cache,sources,storage-local-copies,none} ...]]
                 [--scheduler-greediness SCHEDULER_GREEDINESS]
                 [--scheduler-subsample SCHEDULER_SUBSAMPLE] [--no-hooks]
                 [--debug] [--runtime-profile FILE]
                 [--local-groupid LOCAL_GROUPID] [--attempt ATTEMPT]
                 [--show-failed-logs] [--logger {} [{} ...]]
                 [--job-deploy-sources] [--benchmark-extended]
                 [--container-image IMAGE] [--immediate-submit]
                 [--jobscript SCRIPT] [--jobname NAME] [--flux]
                 [--software-deployment-method {apptainer,conda,env-modules} [{apptainer,conda,env-modules} ...]]
                 [--container-cleanup-images] [--use-conda]
                 [--conda-not-block-search-path-envvars] [--list-conda-envs]
                 [--conda-prefix DIR] [--conda-cleanup-envs]
                 [--conda-cleanup-pkgs [{tarballs,cache}]]
                 [--conda-create-envs-only] [--conda-frontend {conda,mamba}]
                 [--use-apptainer] [--apptainer-prefix DIR]
                 [--apptainer-args ARGS] [--use-envmodules]
                 [--scheduler-solver-path SCHEDULER_SOLVER_PATH]
                 [--deploy-sources QUERY CHECKSUM]
                 [--target-jobs TARGET_JOBS [TARGET_JOBS ...]]
                 [--mode {remote,default,subprocess}]
                 [--report-html-path VALUE]
                 [--report-html-stylesheet-path VALUE]
                 [targets ...]

EXECUTION

targets

Targets to build. May be rules or files.

Default: set()

--dry-run, --dryrun, -n

Do not execute anything, and display what would be done. If you have a very large workflow, use –dry-run –quiet to just print a summary of the DAG of jobs.

Default: False

--profile

Name of profile to use for configuring Snakemake. Snakemake will search for a corresponding folder in /etc/xdg/snakemake and /home/docs/.config/snakemake. Alternatively, this can be an absolute or relative path. The profile folder has to contain a file config.yaml. This file can be used to set default values for command line options in YAML format. For example, –cluster qsub becomes cluster: qsub in the YAML file. Profiles can be obtained from https://github.com/snakemake-profiles. The profile can also be set via the environment variable $SNAKEMAKE_PROFILE. To override this variable and use no profile at all, provide the value none to this argument.

--workflow-profile

Path (relative to current directory) to workflow specific profile folder to use for configuring Snakemake with parameters specific for this workflow (like resources). If this flag is not used, Snakemake will by default use profiles/default if present (searched both relative to current directory and relative to Snakefile, in this order). For skipping any workflow specific profile provide the special value none. Settings made in the workflow profile will override settings made in the general profile (see –profile). The profile folder has to contain a file config.yaml. This file can be used to set default values for command line options in YAML format. For example, –executor slurm becomes executor: slurm in the YAML file. It is advisable to use the workflow profile to set or overwrite e.g. workflow specific resources like the amount of threads of a particular rule or the amount of memory needed. Note that in such cases, the arguments may be given as nested YAML mappings in the profile, e.g. set-threads: myrule: 4 instead of set-threads: myrule=4.

--cache

Store output files of given rules in a central cache given by the environment variable $SNAKEMAKE_OUTPUT_CACHE. Likewise, retrieve output files of the given rules from this cache if they have been created before (by anybody writing to the same cache), instead of actually executing the rules. Output files are identified by hashing all steps, parameters and software stack (conda envs or containers) needed to create them.

--snakefile, -s

The workflow definition in form of a snakefile. Usually, you should not need to specify this. By default, Snakemake will search for Snakefile, snakefile, workflow/Snakefile, workflow/snakefile beneath the current working directory, in this order. Only if you definitely want a different layout, you need to use this parameter.

--cores, -c

Use at most N CPU cores/jobs in parallel. If N is omitted or all, the limit is set to the number of available CPU cores. In case of cluster/cloud execution, this argument sets the maximum number of cores requested from the cluster or cloud scheduler. (See https://snakemake.readthedocs.io/en/stable/snakefiles/rules.html#resources-remote-execution for more info.) This number is available to rules via workflow.cores.

--jobs, -j

Use at most N CPU cluster/cloud jobs in parallel. For local execution this is an alias for –cores (it is though recommended to use –cores in that case). Note: Set to unlimited to allow any number of parallel jobs.

--local-cores

In cluster/cloud mode, use at most N cores of the host machine in parallel (default: number of CPU cores of the host). The cores are used to execute local rules. This option is ignored when not in cluster/cloud mode.

--resources, --res

Define additional resources that shall constrain the scheduling analogously to –cores (see above). A resource is defined as a name and an integer value. E.g. –resources mem_mb=1000. Rules can use resources by defining the resource keyword, e.g. resources: mem_mb=600. If now two rules require 600 of the resource mem_mb they won’t be run in parallel by the scheduler. In cluster/cloud mode, this argument will also constrain the amount of resources requested from the server. (See https://snakemake.readthedocs.io/en/stable/snakefiles/rules.html#resources-remote-execution for more info.)

Default: {}

--set-threads

Overwrite thread usage of rules. This allows to fine-tune workflow parallelization. In particular, this is helpful to target certain cluster nodes by e.g. shifting a rule to use more, or less threads than defined in the workflow. Thereby, THREADS has to be a positive integer, and RULE has to be the name of the rule.

Default: {}

--max-threads

Define a global maximum number of threads available to any rule. Rules requesting more threads (via the threads keyword) will have their values reduced to the maximum. This can be useful when you want to restrict the maximum number of threads without modifying the workflow definition or overwriting rules individually with –set-threads.

--set-resources

Overwrite resource usage of rules. This allows to fine-tune workflow resources. In particular, this is helpful to target certain cluster nodes by e.g. defining a certain partition for a rule, or overriding a temporary directory. Thereby, VALUE has to be a positive integer or a string, RULE has to be the name of the rule, and RESOURCE has to be the name of the resource.

Default: {}

--set-scatter

Overwrite number of scatter items of scattergather processes. This allows to fine-tune workflow parallelization. Thereby, SCATTERITEMS has to be a positive integer, and NAME has to be the name of the scattergather process defined via a scattergather directive in the workflow.

Default: {}

--set-resource-scopes

Overwrite resource scopes. A scope determines how a constraint is reckoned in cluster execution. With RESOURCE=local, a constraint applied to RESOURCE using –resources will be considered the limit for each group submission. With RESOURCE=global, the constraint will apply across all groups cumulatively. By default, only mem_mb and disk_mb are considered local, all other resources are global. This may be modified in the snakefile using the resource_scopes: directive. Note that number of threads, specified via –cores, is always considered local. (See https://snakemake.readthedocs.io/en/stable/snakefiles/rules.html#resources-remote-execution for more info)

Default: {}

--default-resources, --default-res

Define default values of resources for rules that do not define their own values. In addition to plain integers, python expressions over inputsize are allowed (e.g. 2*input.size_mb). The inputsize is the sum of the sizes of all input files of a rule. By default, Snakemake assumes a default for mem_mb, disk_mb, and tmpdir (see below). This option allows to add further defaults (e.g. account and partition for slurm) or to overwrite these default values. The defaults are mem_mb=min(max(2*input.size_mb, 1000), 8000), disk_mb=max(2*input.size_mb, 1000) (i.e., default disk and mem usage is twice the input file size but at least 1GB), and the system temporary directory (as given by $TMPDIR, $TEMP, or $TMP) is used for the tmpdir resource. The tmpdir resource is automatically used by shell commands, scripts and wrappers to store temporary data (as it is mirrored into $TMPDIR, $TEMP, and $TMP for the executed subprocesses). If this argument is not specified at all, Snakemake just uses the tmpdir resource as outlined above.

--preemptible-rules

Define which rules shall use a preemptible machine which can be prematurely killed by e.g. a cloud provider (also called spot instances). This is currently only supported by the Google Life Sciences executor and ignored by all other executors. If no rule names are provided, all rules are considered to be preemptible.

--preemptible-retries

Number of retries that shall be made in order to finish a job from of rule that has been marked as preemptible via the –preemptible-rules setting.

--configfile, --configfiles

Specify or overwrite the config file of the workflow (see the docs). Values specified in JSON or YAML format are available in the global config dictionary inside the workflow. Multiple files overwrite each other in the given order. Thereby missing keys in previous config files are extended by following configfiles. Note that this order also includes a config file defined in the workflow definition itself (which will come first).

Default: []

--config, -C

Set or overwrite values in the workflow config object. The workflow config object is accessible as variable config inside the workflow. Default values can be set by providing a YAML JSON file (see –configfile and Documentation).

--replace-workflow-config

Config files provided via command line do not update and extend the config dictionary of the workflow but instead fully replace it. Keys that are not defined in the provided config files will be undefined even if specified within the workflow config.

Default: False

--envvars

Environment variables to pass to cloud jobs.

Default: set()

--directory, -d

Specify working directory (relative paths in the snakefile will use this as their origin).

--touch, -t

Touch output files (mark them up to date without really changing them) instead of running their commands. This is used to pretend that the rules were executed, in order to fool future invocations of snakemake. Fails if a file does not yet exist. Note that this will only touch files that would otherwise be recreated by Snakemake (e.g. because their input files are newer). For enforcing a touch, combine this with –force, –forceall, or –forcerun. Note however that you lose the provenance information when the files have been created in reality. Hence, this should be used only as a last resort.

Default: False

--keep-going, -k

Go on with independent jobs if a job fails.

Default: False

--rerun-triggers

Possible choices: code, input, mtime, params, software-env

Define what triggers the rerunning of a job. By default, all triggers are used, which guarantees that results are consistent with the workflow code and configuration. If you rather prefer the traditional way of just considering file modification dates, use –rerun-trigger mtime.

Default: ['code', 'input', 'mtime', 'params', 'software-env']

--force, -f

Force the execution of the selected target or the first rule regardless of already created output.

Default: False

--executor, -e

Possible choices: local, dryrun, touch

Specify a custom executor, available via an executor plugin: snakemake_executor_<name>

--forceall, -F

Force the execution of the selected (or the first) rule and all rules it is dependent on regardless of already created output.

Default: False

--forcerun, -R

Force the re-execution or creation of the given rules or files. Use this option if you changed a rule and want to have all its output in your workflow updated.

Default: set()

--consider-ancient

Consider given input items of given rules as ancient, i.e. not triggering re-runs if they are newer than the output files. Putting this into a workflow specific profile (or specifying as argument) allows to overrule rerun triggers caused by file modification dates where the user knows better. RULE is the name of the rule, INPUTITEMS is a comma separated list of input items of the rule (given as name or index (0-based)).

Default: {}

--prioritize, -P

Tell the scheduler to assign creation of given targets (and all their dependencies) highest priority.

Default: set()

--batch

Only create the given BATCH of the input files of the given RULE. This can be used to iteratively run parts of very large workflows. Only the execution plan of the relevant part of the workflow has to be calculated, thereby speeding up DAG computation. It is recommended to provide the most suitable rule for batching when documenting a workflow. It should be some aggregating rule that would be executed only once, and has a large number of input files. For example, it can be a rule that aggregates over samples.

--until, -U

Runs the pipeline until it reaches the specified rules or files. Only runs jobs that are dependencies of the specified rule or files, does not run sibling DAGs.

Default: set()

--omit-from, -O

Prevent the execution or creation of the given rules or files as well as any rules or files that are downstream of these targets in the DAG. Also runs jobs in sibling DAGs that are independent of the rules or files specified here.

Default: set()

--rerun-incomplete, --ri

Re-run all jobs the output of which is recognized as incomplete.

Default: False

--shadow-prefix

Specify a directory in which the shadow directory is created. If not supplied, the value is set to the .snakemake directory relative to the working directory.

--strict-dag-evaluation

Possible choices: cyclic-graph, functions, periodic-wildcards

Strict evaluation of rules’ correctness even when not required to produce the output files.

Default: set()

--scheduler

Possible choices: ilp, greedy

Specifies if jobs are selected by a greedy algorithm or by solving an ilp. The ilp scheduler aims to reduce runtime and hdd usage by best possible use of resources.

Default: 'greedy'

--wms-monitor

IP and port of workflow management system to monitor the execution of snakemake (e.g. http://127.0.0.1:5000) Note that if your service requires an authorization token, you must export WMS_MONITOR_TOKEN in the environment.

--wms-monitor-arg

If the workflow management service accepts extra arguments, provide. them in key value pairs with –wms-monitor-arg. For example, to run an existing workflow using a wms monitor, you can provide the pair id=12345 and the arguments will be provided to the endpoint to first interact with the workflow

--scheduler-ilp-solver

Possible choices: PULP_CBC_CMD

Specifies solver to be utilized when selecting ilp-scheduler.

Default: 'COIN_CMD'

--conda-base-path

Path of conda base installation (home of conda, mamba, activate) (internal use only).

--no-subworkflows, --nosw

Do not evaluate or execute subworkflows.

Default: False

--precommand

Only used in case of remote execution. Command to be executed before Snakemake executes each job on the remote compute node.

--dag

Possible choices: dot, mermaid-js

Do not execute anything and print the directed acyclic graph of jobs in the dot language or in mermaid-js. Recommended use on Unix systems: snakemake –dag | dot | display. Note print statements in your Snakefile may interfere with visualization.

GROUPING

--groups

Assign rules to groups (this overwrites any group definitions from the workflow).

Default: {}

--group-components

Set the number of connected components a group is allowed to span. By default, this is 1, but this flag allows to extend this. This can be used to run e.g. 3 jobs of the same rule in the same group, although they are not connected. It can be helpful for putting together many small jobs or benefitting of shared memory setups.

Default: {}

REPORTS

--report

Create a self-contained HTML report with default statistics, provenance information and user-specified results. For smaller datasets with a limited report complexity, you can specify an .html file and all results will be embedded directly into this file. For customized reports on larger sample sizes, it makes more sense to specify a .zip file. The resulting archive will spread the contents across a folder structure, for a quicker loading of individual results. You can unpack this archive anywhere and open the report.html file in its main folder to view the report in any web browser.

--report-after-run

After finishing the workflow, directly create the report. It is required to provide –report.

Default: False

--report-stylesheet

Custom stylesheet to use for report. In particular, this can be used for branding the report with e.g. a custom logo, see docs.

--reporter

Specify a custom report plugin. By default, Snakemake’s builtin html reporter will be used. For custom reporters, check out their command line options starting with –report-.

NOTEBOOKS

--draft-notebook

Draft a skeleton notebook for the rule used to generate the given target file. This notebook can then be opened in a jupyter server, executed and implemented until ready. After saving, it will automatically be reused in non-interactive mode by Snakemake for subsequent jobs.

--edit-notebook

Interactively edit the notebook associated with the rule used to generate the given target file. This will start a local jupyter notebook server. Any changes to the notebook should be saved, and the server has to be stopped by closing the notebook and hitting the Quit button on the jupyter dashboard. Afterwards, the updated notebook will be automatically stored in the path defined in the rule. If the notebook is not yet present, this will create an empty draft.

--notebook-listen

The IP address and PORT the notebook server used for editing the notebook (–edit-notebook) will listen on.

Default: 'localhost:8888'

UTILITIES

--lint

Possible choices: text, json

Perform linting on the given workflow. This will print snakemake specific suggestions to improve code quality (work in progress, more lints to be added in the future). If no argument is provided, plain text output is used.

--generate-unit-tests

Automatically generate unit tests for each workflow rule. This assumes that all input files of each job are already present. Jobs without present input files will be skipped (a warning will be issued). For each rule, one test case will be created and, after successful execution, tests can be run with pytest TESTPATH.

--containerize

Print a Dockerfile that provides an execution environment for the workflow, including all conda environments.

Default: False

--export-cwl

Compile workflow to CWL and store it in given FILE.

--list-rules, --list, -l

Show available rules in given Snakefile.

Default: False

--list-target-rules, --lt

Show available target rules in given Snakefile.

Default: False

--rulegraph

Possible choices: dot, mermaid-js

Do not execute anything and print the dependency graph of rules in the dot language or in mermaid-js. This will be less crowded than above DAG of jobs, but also show less information. Note that each rule is displayed once, hence the displayed graph will be cyclic if a rule appears in several steps of the workflow. Use this if above option leads to a DAG that is too large. Recommended use on Unix systems: snakemake –rulegraph | dot | display. Note print statements in your Snakefile may interfere with visualization.

--filegraph

Do not execute anything and print the dependency graph of rules with their input and output files in the dot language. This is an intermediate solution between above DAG of jobs and the rule graph. Note that each rule is displayed once, hence the displayed graph will be cyclic if a rule appears in several steps of the workflow. Use this if above option leads to a DAG that is too large. Recommended use on Unix systems: snakemake –filegraph | dot | display. Note print statements in your Snakefile may interfere with visualization.

Default: False

--d3dag

Print the DAG in D3.js compatible JSON format.

Default: False

--summary, -S

Print a summary of all files created by the workflow. The has the following columns: filename, modification time, rule version, status, plan. Thereby rule version contains the version the file was created with (see the version keyword of rules), and status denotes whether the file is missing, its input files are newer or if version or implementation of the rule changed since file creation. Finally the last column denotes whether the file will be updated or created during the next workflow execution.

Default: False

--detailed-summary, -D

Print a summary of all files created by the workflow. The has the following columns: filename, modification time, rule version, input file(s), shell command, status, plan. Thereby rule version contains the version the file was created with (see the version keyword of rules), and status denotes whether the file is missing, its input files are newer or if version or implementation of the rule changed since file creation. The input file and shell command columns are self explanatory. Finally the last column denotes whether the file will be updated or created during the next workflow execution.

Default: False

--archive

Archive the workflow into the given tar archive FILE. The archive will be created such that the workflow can be re-executed on a vanilla system. The function needs conda and git to be installed. It will archive every file that is under git version control. Note that it is best practice to have the Snakefile, config files, and scripts under version control. Hence, they will be included in the archive. Further, it will add input files that are not generated by by the workflow itself and conda environments. Note that symlinks are dereferenced. Supported formats are .tar, .tar.gz, .tar.bz2 and .tar.xz.

--cleanup-metadata, --cm

Cleanup the metadata of given files. That means that snakemake removes any tracked version info, and any marks that files are incomplete.

--cleanup-shadow

Cleanup old shadow directories which have not been deleted due to failures or power loss.

Default: False

--skip-script-cleanup

Don’t delete wrapper scripts used for execution

Default: False

--unlock

Remove a lock on the working directory.

Default: False

--list-changes, --lc

Possible choices: params, code, input

List all output files for which the given items (code, input, params) have changed since creation.

--list-input-changes, --li

List all output files for which the defined input files have changed in the Snakefile (e.g. new input files were added in the rule definition or files were renamed). For listing input file modification in the filesystem, use –summary.

Default: False

--list-params-changes, --lp

List all output files for which the defined params have changed in the Snakefile.

Default: False

--list-untracked, --lu

List all files in the working directory that are not used in the workflow. This can be used e.g. for identifying leftover files. Hidden files and directories are ignored.

Default: False

--delete-all-output

Remove all files generated by the workflow. Use together with –dry-run to list files without actually deleting anything. Note that this will not recurse into subworkflows. Write-protected files are not removed. Nevertheless, use with care!

Default: False

--delete-temp-output

Remove all temporary files generated by the workflow. Use together with –dry-run to list files without actually deleting anything. Note that this will not recurse into subworkflows.

Default: False

--keep-incomplete

Do not remove incomplete output files by failed jobs.

Default: False

--drop-metadata

Drop metadata file tracking information after job finishes. Provenance-information based reports (e.g. –report and the –list_x_changes functions) will be empty or incomplete.

Default: False

--version, -v

show program’s version number and exit

OUTPUT

--printshellcmds, -p

Print out the shell commands that will be executed.

Default: False

--debug-dag

Print candidate and selected jobs (including their wildcards) while inferring DAG. This can help to debug unexpected DAG topology or errors.

Default: False

--nocolor

Do not use a colored output.

Default: False

--quiet, -q

Possible choices: all, host, progress, reason, rules

Do not output certain information. If used without arguments, do not output any progress or rule information. Defining all results in no information being printed at all.

--print-compilation

Print the python representation of the workflow.

Default: False

--verbose

Print debugging output.

Default: False

BEHAVIOR

--force-use-threads

Force threads rather than processes. Helpful if shared memory (/dev/shm) is full or unavailable.

Default: False

--allow-ambiguity, -a

Don’t check for ambiguous rules and simply use the first if several can produce the same file. This allows the user to prioritize rules by their order in the snakefile.

Default: False

--nolock

Do not lock the working directory

Default: False

--ignore-incomplete, --ii

Do not check for incomplete output files.

Default: False

--max-inventory-time

Spend at most SECONDS seconds to create a file inventory for the working directory. The inventory vastly speeds up file modification and existence checks when computing which jobs need to be executed. However, creating the inventory itself can be slow, e.g. on network file systems. Hence, we do not spend more than a given amount of time and fall back to individual checks for the rest.

Default: 20

--trust-io-cache

Tell Snakemake to assume that all input and output file existence and modification time queries performed in previous dryruns are still valid and therefore don’t have to be repeated. This can lead to speed-ups, but implies that input and output have not been modified manually in between. Non dry-run execution will automatically invalidate the cache and lead to redoing the queries.

Default: False

--max-checksum-file-size

Compute the checksum during DAG computation and job postprocessing only for files that are smaller than the provided threshold (given in any valid size unit, e.g. 1MB, which is also the default).

Default: 1000000

--latency-wait, --output-wait, -w

Wait given seconds if an output file of a job is not present after the job finished. This helps if your filesystem suffers from latency.

Default: 5

--wait-for-free-local-storage

Wait for given timespan for enough free local storage when downloading remote storage files. If not set, no waiting is performed.

--wait-for-files

Wait –latency-wait seconds for these files to be present before executing the workflow. This option is used internally to handle filesystem latency in cluster environments.

--wait-for-files-file

Same behaviour as –wait-for-files, but file list is stored in file instead of being passed on the commandline. This is useful when the list of files is too long to be passed on the commandline.

--queue-input-wait-time

Set the interval in seconds to check for new input in rules that use from_queue to obtain input files.

Default: 10

--notemp, --nt

Ignore temp() declarations. This is useful when running only a part of the workflow, since temp() would lead to deletion of probably needed files by other parts of the workflow.

Default: False

--all-temp

Mark all output files as temp files. This can be useful for CI testing, in order to save space.

Default: False

--unneeded-temp-files

Given files will not be uploaded to storage and immediately deleted after job or group job completion.

Default: frozenset()

--keep-storage-local-copies

Keep local copies of remote input and output files.

Default: False

--not-retrieve-storage

Do not retrieve remote files (default is to retrieve remote files).

Default: False

--target-files-omit-workdir-adjustment

Do not adjust the paths of given target files relative to the working directory.

Default: False

--allowed-rules

Only consider given rules. If omitted, all rules in Snakefile are used. Note that this is intended primarily for internal use and may lead to unexpected results otherwise.

--max-jobs-per-timespan

Maximal number of job submissions/executions per timespan. Format: <number><timespan>, e.g. 50/1m or 0.5/1s.

Default: 100/1s

--max-jobs-per-second

Maximal number of job submissions/executions per second. Deprecated in favor of –max-jobs-per-timespan.

--max-status-checks-per-second

Maximal number of job status checks per second; fractions allowed.

Default: 10

--seconds-between-status-checks

Number of seconds to wait between two rounds of status checks.

Default: 10

--retries, --restart-times, -T

Number of times to restart failing jobs.

Default: 0

--wrapper-prefix

URL prefix for wrapper directive. Set this to use your fork or a local clone of the repository, e.g., use a git URL like git+file://path/to/your/local/clone@.

Default: 'https://github.com/snakemake/snakemake-wrappers/raw/'

--default-storage-provider

Specify default storage provider to be used for all input and output files that don’t yet specify one (e.g. s3). See https://snakemake.github.io/snakemake-plugin-catalog for available storage provider plugins. If not set or explicitly none, no default storage provider will be used.

--default-storage-prefix

Specify prefix for default storage provider. E.g. a bucket name.

Default: ''

--local-storage-prefix

Specify prefix for storing local copies of storage files and folders (e.g. local scratch disk). Environment variables will be expanded.

Default: .snakemake/storage

--remote-job-local-storage-prefix

Specify prefix for storing local copies of storage files and folders (e.g. local scratch disk) in case of remote jobs (e.g. cluster or cloud jobs). Environment variables will be expanded within the remote job.

Default: .snakemake/storage

--shared-fs-usage

Possible choices: input-output, persistence, software-deployment, source-cache, sources, storage-local-copies, none

Set assumptions on shared filesystem for non-local workflow execution. To disable any sharing via the filesystem, specify none. Usually, the executor plugin sets this to a correct default. However, sometimes it is worth tuning this setting, e.g. for optimizing cluster performance. For example, when using –default-storage-provider fs together with a cluster executor like slurm, you might want to set –shared-fs-usage persistence software-deployment sources source-cache, such that software deployment and data provenance will be handled by NFS but input and output files will be handled exclusively by the storage provider.

Default: frozenset({<SharedFSUsage.SOFTWARE_DEPLOYMENT: 2>, <SharedFSUsage.PERSISTENCE: 0>, <SharedFSUsage.STORAGE_LOCAL_COPIES: 4>, <SharedFSUsage.SOURCES: 3>, <SharedFSUsage.SOURCE_CACHE: 5>, <SharedFSUsage.INPUT_OUTPUT: 1>})

--scheduler-greediness, --greediness

Set the greediness of scheduling. This value between 0 and 1 determines how careful jobs are selected for execution. The default value (1.0) provides the best speed and still acceptable scheduling quality.

Default: 1.0

--scheduler-subsample

Set the number of jobs to be considered for scheduling. If number of ready jobs is greater than this value, this number of jobs is randomly chosen for scheduling; if number of ready jobs is lower, this option has no effect. This can be useful on very large DAGs, where the scheduler can take some time selecting which jobs to run.

--no-hooks

Do not invoke onstart, onsuccess or onerror hooks after execution.

Default: False

--debug

Allow to debug rules with e.g. PDB. This flag allows to set breakpoints in run blocks.

Default: False

--runtime-profile

Profile Snakemake and write the output to FILE. This requires yappi to be installed.

--local-groupid

Internal use only: Name for local groupid.

Default: 'local'

--attempt

Internal use only: define the initial value of the attempt parameter.

Default: 1

--show-failed-logs

Automatically display logs of failed jobs.

Default: False

--logger

Specify one or more custom loggers, available via logger plugins.

Default: []

--job-deploy-sources

Whether the workflow sources shall be deployed before a remote job is started. Only applies if –no-shared-fs is set or executors are used that imply no shared FS (e.g. the kubernetes executor).

Default: False

--benchmark-extended

Write extended benchmarking metrics.

Default: False

REMOTE EXECUTION

--container-image

Docker image to use, e.g., when submitting jobs to kubernetes. Defaults to https://hub.docker.com/r/snakemake/snakemake, tagged with the same version as the currently running Snakemake instance. Note that overwriting this value is up to your responsibility. Any used image has to contain a working snakemake installation that is compatible with (or ideally the same as) the currently running version.

Default: 'snakemake/snakemake:v9.1.2.dev3'

--immediate-submit, --is

Immediately submit all jobs to the cluster instead of waiting for present input files. This will fail, unless you make the cluster aware of job dependencies, e.g. via: $ snakemake –cluster ‘sbatch –dependency {dependencies}’. Assuming that your submit script (here sbatch) outputs the generated job id to the first stdout line, {dependencies} will be filled with space separated job ids this job depends on. Does not work for workflows that contain checkpoint rules.

Default: False

--jobscript, --js

Provide a custom job script for submission to the cluster. The default script resides as jobscript.sh in the installation directory.

--jobname, --jn

Provide a custom name for the jobscript that is submitted to the cluster (see –cluster). The wildcard {jobid} has to be present in the name.

Default: 'snakejob.{name}.{jobid}.sh'

FLUX

--flux

Execute your workflow on a flux cluster. Flux can work with both a shared network filesystem (like NFS) or without. If you don’t have a shared filesystem, additionally specify –no-shared-fs.

Default: False

SOFTWARE DEPLOYMENT

--software-deployment-method, --deployment-method, --deployment, --sdm

Possible choices: apptainer, conda, env-modules

Specify software environment deployment method.

Default: set()

--container-cleanup-images

Remove unused containers

Default: False

CONDA

--use-conda

If defined in the rule, run job in a conda environment. If this flag is not set, the conda directive is ignored.

Default: False

--conda-not-block-search-path-envvars

Do not block environment variables that modify the search path (R_LIBS, PYTHONPATH, PERL5LIB, PERLLIB) when using conda environments.

Default: False

--list-conda-envs

List all conda environments and their location on disk.

Default: False

--conda-prefix

Specify a directory in which the conda and conda-archive directories are created. These are used to store conda environments and their archives, respectively. If not supplied, the value is set to the .snakemake directory relative to the invocation directory. If supplied, the –use-conda flag must also be set. The value may be given as a relative path, which will be extrapolated to the invocation directory, or as an absolute path. The value can also be provided via the environment variable $SNAKEMAKE_CONDA_PREFIX. In any case, the prefix may contain environment variables which will be properly expanded. Note that if you use remote execution e.g. on a cluster and you have node specific values for this, you should disable assuming shared fs for software-deployment (see –shared-fs-usage).

--conda-cleanup-envs

Cleanup unused conda environments.

Default: False

--conda-cleanup-pkgs

Possible choices: tarballs, cache

Cleanup conda packages after creating environments. In case of tarballs mode, will clean up all downloaded package tarballs. In case of cache mode, will additionally clean up unused package caches.

Default: tarballs

--conda-create-envs-only

If specified, only creates the job-specific conda environments then exits. The –use-conda flag must also be set.

Default: False

--conda-frontend

Possible choices: conda, mamba

Choose the conda frontend for installing environments.

Default: 'conda'

APPTAINER/SINGULARITY

--use-apptainer, --use-singularity

If defined in the rule, run job within a apptainer/singularity container. If this flag is not set, the singularity directive is ignored.

Default: False

--apptainer-prefix, --singularity-prefix

Specify a directory in which apptainer/singularity images will be stored.If not supplied, the value is set to the .snakemake directory relative to the invocation directory. If supplied, the –use-apptainer flag must also be set. The value may be given as a relative path, which will be extrapolated to the invocation directory, or as an absolute path. If not supplied, APPTAINER_CACHEDIR is used. In any case, the prefix may contain environment variables which will be properly expanded. Note that if you use remote execution e.g. on a cluster and you have node specific values for this, you should disable assuming shared fs for software-deployment (see –shared-fs-usage).

--apptainer-args, --singularity-args

Pass additional args to apptainer/singularity.

Default: ''

ENVIRONMENT MODULES

--use-envmodules

If defined in the rule, run job within the given environment modules, loaded in the given order. This can be combined with –use-conda and –use-singularity, which will then be only used as a fallback for rules which don’t define environment modules.

Default: False

INTERNAL

--scheduler-solver-path

Internal use only: Set the PATH to search for scheduler solver binaries.

--deploy-sources

Internal use only: Deploy sources archive from given storage provider query to the current working subdirectory and control for archive checksum to proceed.

--target-jobs

Internal use only: Target particular jobs by RULE:WILDCARD1=VALUE,WILDCARD2=VALUE,…

Default: set()

--mode

Possible choices: remote, default, subprocess

Internal use only: Set execution mode of Snakemake.

Default: default

html executor settings

--report-html-path

Path to the report file (either .html or .zip). Use zip if your report contains large results or directories with htmlindex as results.

Default: <dataclasses._MISSING_TYPE object at 0x77a1b52a8110>

--report-html-stylesheet-path

Path to a custom stylesheet for the report.

Default: <dataclasses._MISSING_TYPE object at 0x77a1b52a8110>