(Sampler_Brito) Add sampler

[jsbrittain]

Add the Anderson Brito subsampler (https://github.com/andersonbrito/subsampler)

To test the module:

• Clone the originating repo (git@github.com:jsbrittain/vneyard.git) and checkout this branch (git checkout sampler_brito) [this will not be necessary once the module is added to the repository].
• Open GRAPEVNE and provide the repo base path (where you cloned vneyard/) as a local repository.
• Drag and drop Subsampler_Brito into the scene from the EPI project. Link a local folder as input using LinkLocalFolder (from the Utility project). You can use the sample data provided in vneyard/workflows/EPI/modules/Subsampler_Brito/results/data; do the same with the config folder (note that if you remove the subsampler config filenames from the module then it will run with defaults).
• Build and Run the workflow with the existing default options (or modify and build).
• Open Results to see the output; see below for a screenshot of the workflow with results.

Screenshot:

resolves #2

---

Improvements that can be made:

• [x] Make config input optional (i.e. if no config files found, use some defaults)
• [x] Docstring can likely be improved with domain-specific knowledge

[jsbrittain]

Docstring currently reads:

Subsampler (Anderson Brito)

A pipeline for subsampling genomic data based on epidemiological time series data. This pipeline was developed by Anderson Brito and is available at: https://github.com/andersonbrito/subsampler/tree/master.

Ports: data: Input data (metadata and case data) config: Configuration files (keep, remove and filter files), default none if unconnected

Params: Metadata file (str): Name of the metadata file Case data file (str): Name of the case data file Subsampler: keep file (str): File designating sequences to keep. This should be a plain-text (.txt) file containing one ID (e.g. EPI_ISL_402125) per line. Leave blank for none. remove file (str): File designating sequences to remove. This should be a plain-text (.txt) file containing one ID (e.g. EPI_ISL_12345678) per line. Leave blank for none. filter file (str): File containing sequence filters. This should be a tab-delimited file (.tsv) with columns action, column and value. action can contain "exclude", the column should contain the name of the column to filter on, and the value column should contain the value to filter on. Leave blank for none. ID Column (str): Column name for the ID GEO Column (str): Column name for the geographic location Date column (str): Column name for the date Baseline (decimal): Reference genome size (int): Reference genome size Max missing (int): Maximum number of missing data Seed number (int): Seed Start date (str): Start date End date (str): End date Unit (str): Time unit

[jsbrittain]

@joetsui1994 Could you take a look at the docstring in particular to ensure that it provideds sufficient information for the end user to run the module. Thanks!

[joetsui1994]

@jsbrittain Looks great! I just tested it and it works.

I would suggest a few changes in the docstring: 1) Give either an example for start/end date, or the required format (seems to be YYYY-MM-DD)? 2) What are the options for time unit? (seems to be ['week', 'month', 'year', 'full']) 3) change "Baseline" to "Baseline sampling proportion" 4) "Max" to "Max."

Otherwise this looks good to me!