ggd predict-path

[Click here to return to the home page]

ggd predict-path is used to predict that path of a data file from a ggd data package that has not been installed yet. Workflows, such as Snakemake, require knowledge of the file path prior to running the workflow. Using the predict-path tool would allow someone using Snakemake to provide the predict file paths of the data files they would like to use without having to pre-install the data packages.The installation of the data packages could take place later in the workflow process.

Using ggd predict-path

Use ggd predict-path to predict the path of a data file prior to installing the ggd data package. Running ggd predict-path -h will give you the following message:

predict-path arguments:

ggd predict-path	Get a predicted install file path for a data package before it is installed. (Use for workflows, such as Snakemake)
-h, --help	show this help message and exit
-c, --channel	The ggd channel of the recipe to fine. (Deafult = genomics)
--prefix	(Optional) The name or the full directory path to an conda environment. The predicted path will be based on this conda environment. When installing, the data package should also be installed in this environment. (Only needed if not predicting for a path in the current conda environment)
--id, meta-recipe ID	(Optional) The ID to predict the path for if the package is a meta-recipe. If it is not a meta-recipe it will be ignored
-pn, --package-name	Required The name of the data package to predict a file path for
-fn, --file-name	Required if –dir-path not used The name of the file to predict that path for. It is best if you give the full and correct name of the file to predict the path for. If not, ggd will try to identify the right file, but won’t guarantee that it is the right file
-dir-path	Required if –file-name not used Whether or not to get the predicted directory path rather then the predicted file path. If both –file-name and –dir-path are provided the –file-name will be used and –dir-path will be ignored

Additional argument explanation:

Required arguments:

• -pn: The -pn flag represents the data package name for which to predict a file path for. (NOTE: the file path is for a data file from this data package and not the data package itself)

One Argument Required:

• -fn: The -fn flag represents the name of the data file for which to predict the file path for. (NOTE: this is the actual file from the data package to predict the path for) You can use ggd search or check the Available packages page to get the name of the files provided by the package

• –dir-path The --dir-path flag is used to the the predicted directory path of the installed files rather then the exact file path. This is specifically useful for meta-recipes, where the file-path cannot be predicted because of the dynamic nature of meta-recipes.

Note

Either the --file-name or the --dir-path parameter is required. If both are provided then the --file-path will be used and the --dir-path ignored. If neither one is given then the predict-path command will exit and prompt for one.

Optional arguments:

• -c: The -c flag represents the ggd channel. The default channel is genomics.

• –prefix: The --prefix flag is used to get to the install path for a data file in a specific conda environment/prefix that is different then the current environment.

• –id The --id flag is used to for predicting the directory path of an ID specifc meta-recipe. It is not possible to predict the file path of a meta-recipe due to the dynamic nature of the recipe iteself, however, it is possible to predict the directory path. An ID is required when install a meta-recipe, and this ID represents the ID that would be used when install a desired meta-recipe. This ID will be ignored if the path being predicted is NOT for a meta-recipe. Additionaly, if this parameter is not used then the directory path to the non ID specifc meta-recipe will be provided, which is of no use.

Examples

1. Predict the path of the chromsize file:

$ ggd predict-path -pn hg19-chromsizes-ggd-v1 -fn hg19-chromsizes-ggd-v1.txt

  <conda root>/share/ggd/Homo_sapiens/hg19/hg19-chromsizes-ggd-v1/1/hg19-chromsizes-ggd-v1.txt

2. Predict the path of the chromsize file in the “data” environment:

$ ggd predict-path -pn hg19-chromsizes-ggd-v1 -fn hg19-chromsizes-ggd-v1.txt --prefix data

  <data environment>/share/ggd/Homo_sapiens/hg19/hg19-chromsizes-ggd-v1/1/hg19-chromsizes-ggd-v1.txt

3. Predict the directory path of a GEO meta-recipe:

$ ggd predict-path --package-name meta-recipe-geo-accession-geo-v1 --dir-path --id GSE123

 <conda root>/share/ggd/meta-recipe/meta-recipe/gse123-geo-v1/1/

Note

If you are getting the predicted path for a meta-recipe you need to add the --id parameter, otherwise, you will get the directory path to a main meta-recipe, which should never contain any data files. (The wrong path)