Using the Rxnat Package


knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
library(Rxnat)

What is XNAT?

XNAT is an open source imaging informatics platform developed by the Neuroinformatics Research Group at Washington University. XNAT was originally developed in the Buckner Lab at Washington University, now at Harvard University. It facilitates common management, productivity, and quality assurance tasks for imaging and associated data. Thanks to its extensibility, XNAT can be used to support a wide range of imaging-based projects.

Who is using XNAT?

There are several projects that currently use XNAT to manage a vast number of datasets.

For a more complete list of XNAT implementations around the world you can click here.

Installing the Rxnat package

You can install Rxnat from github with: ``` {r, eval = FALSE}

install.packages("devtools")

library(devtools) devtools::install_github("adigherman/Rxnat")

## Accessing XNAT Data

### How to get a username/password for a XNAT project
XNAT projects can be public or private. In order to access a private repository a set of credentials are required. To obtain a user name and password combo you will need to visit:

* HCP (Human Connectome Project) - [new account request](https://db.humanconnectome.org) and click on the Register button
* NITRC - [new account request](https://www.nitrc.org/account/register.php). Some of the NITRC hosted projects will require additional access requests but this can be easily requested using the NITRC web interface for each project.
* XNAT Central - this is mostly public, but you can also perform a [new account request](https://central.xnat.org/app/template/Register.vm) if any of the projects require it. 

### Establishing a XNAT connection with the `Rxnat` package
The `Rxnat` package will accept credentials provided in the function call or read from the system environment.

#### Function parameters
To establish a connection using the credentials as function parameters we can call the `xnat_connect` function:
``` {r, eval = FALSE}
nitrc <- xnat_connect('https://nitrc.org/ir', username='XXXX', password='YYYY', xnat_name=NULL)

Setting up system environment variables

To use system environment variables we need to add them to the .Renviron file located in the user's home folder. Use Sys.getenv("HOME") to get the path (for unix/osx users the location can be easily accessed with ~, eg. vi ~/.Renviron). The Rxnat package will be able to automatically read / use a system environment variable provided the following format is used: XXXX_Rxnat_USER and XXXX_Rxnat_PASS. XXXX is provided as an argument when an XNAT connection is initiated.

As an example NITRC is used as argument and the system environment variables names should be NITRC_Rxnat_USER, and NITRC_Rxnat_PASS. ``` {r, eval = FALSE} nitrc <- xnat_connect('https://nitrc.org/ir', xnat_name='NITRC')

NITRC has some public image repositories, a username and password are not required to connect/query them. To create a connection to the public NITRC resources we call the `xnat_connect` function.

``` {r}
nitrc <- xnat_connect('https://nitrc.org/ir')

Get list of available XNAT projects

Once a connection is established using the xnat_connect function a list of available projects can be easily retrieved by using the class internal function projects:

nitrc_projects <- nitrc$projects()
head(nitrc_projects[c('ID','name')])

Retrieve a list with all accessible subjects

A full list of subjects for each XNAT connection can be retrieved using the subjects function:

nitrc_subjects <- nitrc$subjects()
head(nitrc_subjects)

Get full list of experiments

To obtain a full list of experiments the experiments function will be used:

nitrc_experiments <- nitrc$experiments()
head(nitrc_experiments)

Get the complete list of resources for a specific experiment

The scan resources for an experiment can be retrieved using the get_xnat_experiment_resources function:

xnat_E00026_resources <- nitrc$get_xnat_experiment_resources('xnat_E00026')
head(xnat_E00026_resources[c('Name','URI')])

Query the XNAT projects for matching entries

If you are interested just in a subset of subjects/experiments that match a certain criteria you can use the query_scan_resources function. Accepted query parameters are:

To retrieve a list of all subject IDs and associated experiment IDs we can use the query_scan_resources function. In the example below, we are querying the HCP XNAT database for all subjects belonging the the HCP_500 project with scans taken at age 26.

nitrc_age_26 <- query_scan_resources(nitrc,age='26')
head(nitrc_age_26[c("subject_ID","experiment_ID", "Project", "Age")])

Getting Data: Download a single scan image/resource file

To download a single file we will use the download_file function. Using the first experiment_ID from the above example, we will get all scan resources associated with it first.

``` {r, eval = FALSE} scan_resources <- get_scan_resources(nitrc,'xnat_E00026') scan_resources[1,c("Name","URI")]

To download the resource file (100307_3T_BIAS_BC.nii.gz) we will do:

``` {r, eval = FALSE}
> download_xnat_file(nitrc,"/data/experiments/xnat_E00026/scans/anat_mprage_anonymized/resources/3967/files/scan_mprage_anonymized.nii.gz", file_dir = tempdir())
[1] "/var/folders/kr/05bm5krj0r3fpwxfdmx4xthm0000gn/T//RtmpTNSN5E/scan_mprage_anonymized.nii.gz"

Getting Data: Download a directory of data

To download all the T1 type images from experiment ConnectomeDB_E03657 we will use the download_xnat_dir function.

{r, eval = FALSE} download_xnat_dir(nitrc, experiment_ID='NITRC_IR_E10507',scan_type='T2', file_dir = tempdir(), verbose=TRUE) /var/folders/kr/05bm5krj0r3fpwxfdmx4xthm0000gn/T//RtmpTNSN5E/NITRC_IR_E10507.zip Downloading: 5.2 MB [1] "/var/folders/kr/05bm5krj0r3fpwxfdmx4xthm0000gn/T//RtmpTNSN5E/NITRC_IR_E10507.zip"



Try the Rxnat package in your browser

Any scripts or data that you put into this service are public.

Rxnat documentation built on Sept. 16, 2020, 1:08 a.m.