taigr

Light weight R package for loading datasets from taiga. Conveniently caches data to your hard disk.

The latest version of this library is a thin wrapper around taigaclient which is part of taigapy. As a result the first step is to install taigapy into a python environment.

Prerequistes:

1. conda: You can download miniconda from https://docs.anaconda.com/free/miniconda/miniconda-install/ and the instructions below will use the conda command to set up the python environment in which we'll install taigapy.
2. git: Can be downloaded from https://git-scm.com/downloads and is used by some of the steps below to pull code from github.
3. The R devtools package: Can be installed via install.packages('devtools')

Installation

1. Create a python environment and installing taigapy into it:

conda create -n taigapy python=3.9
conda activate taigapy
pip install git+https://github.com/broadinstitute/taigapy.git

2. Install taigr in an R session

library(devtools)
devtools::install_github("https://github.com/broadinstitute/taigr")

3. Set up token

You need to get your authorization token so the client library can make requests on your behalf. Go to https://cds.team/taiga/token/ and click on the "Copy" button to copy your token. Paste your token in a file at ~/.taiga/token. You can do this in R via:

On MacOS:

dir.create(path.expand("~/.taiga"))
write("YOUR_TOKEN_HERE", file=path.expand("~/.taiga/token"))

On Windows (write it to two locations to be safest):

token <- "YOUR_TOKEN_HERE"
home.dir <- Sys.getenv("USERPROFILE")
dir.create(file.path(home.dir, ".taiga"))
write(token, file=file.path(home.dir, ".taiga", "token"))
home.dir <- Sys.getenv("HOME")
dir.create(file.path(home.dir, ".taiga"))
write(token, file=file.path(home.dir, ".taiga", "token"))

4. Set up path to taigaclient

taigr needs to know where to file the taigaclient executable that got installed when taigapy was installed. You can tell it by setting the taigaclient.path option in R.

In MacOS, this would typically be:

options(taigaclient.path=path.expand("~/miniconda3/envs/taigapy/bin/taigaclient"))

Or in Windows, this would typically be:

options(taigaclient.path=file.path(Sys.getenv("USERPROFILE"), "miniconda3/envs/taigapy/Scripts/taigaclient"))

( You may want to put this command in your ".Rprofile" so that it automatically executes each time R starts up as described in https://www.statmethods.net/interface/customizing.html )

At this point taigr should be successfully set up and you can test it out by running:

taigr::load.from.taiga("taigr-data-40f2.7/tiny_matrix")

( If you get an error like The following command reported an error... the first thing to check is whether the command it's reporting failed does actually exist at that location. )

If successful, it should have retreived the 2x3 matrix which was listed at https://cds.team/taiga/dataset/taigr-data-40f2/7

Installation in a docker container

Similarly, install taigapy and then install taigr from a source bundle to avoid needing to install devtools

RUN curl -L -o taigr.zip https://github.com/broadinstitute/taigr/archive/master.zip && \
    unzip taigr.zip && \
    cd taigr-master && \
    R CMD INSTALL .

Quick start

library(taigr)
demeter <- load.from.taiga(
	data.name = "demeter-2-data-20123df",
	data.version = 1)

Package documentation

package?taigr

?load.from.taiga

Development

To test your changes, run the following line in R

devtools::install(pkg=".")