The Ontario Neurodegenerative Disease Research Initiative (ONDRI) is a series of studies of neurodegenerative and cerebrovascular diseases. For an overview of the foundational prospective, observational, longitudinal study at baseline see this preprint. For more general information on the project, visit ondri.ca, and follow the study on Twitter: @ONDRISTUDY.

ondri-nibs

The ondri-nibs page is a github organization page for the Ontario Neurodegenerative Disease Research Initiative’s (ONDRI) Neuroinformatics & Biostatistics (NIBS) team to house materials for public (and internal) consumption.

Generally, this organization page has multiple repositories, each repository contains a unique item (e.g., Shiny apps, packages) or conceptual content (e.g., documentation). This page will provide links to each of those as the project grows and changes.

To note, all materials here are licensed as follows unless otherwise stated:

• GPL-3 for software

• CC-BY for documentation

If you use or adapt any materials here, please ensure you correctly do so according to their licenses and with full attribution.

contributors

This repository is a single file to keep track of all contributors to the NIBS material provided here.

ondridocs

This repository is a home for all the external facing (and some internally useful) documentation.

toy data

This repository provides two “toy” data sets. These data sets help illustrate a few things:

• Each one shows how a tabular and non-tabular data set would be constructed

• The data are synthetic, and also provide an illustrative data set for use with all of the apps and methods here (e.g., standards, data preparation, outlier detection)

• See also this video for an explanation of the standards we implemented for the data

ONDRI data.frames

This repository is a simple package with one primary function ONDRI_df(). Its goal is to read in a DATA.csv and DICT.csv that are part of the ONDRI data packages, and automatically handle some of the specifics about our data packages. That includes:

• Recoding and preserving the different missing data, as defined in the ONDRI standards.

• Include information from the dictionary—such as DATA_TYPES—as part of a data.frame.

• Some printing utilities to help see ONDRI’s data types and ONDRI’s missing coded data in the data packages

• Some simple subsetting functions based on data types.

ondricolors

This repository houses a lightweight R package designed with one purpose: to provide a color palette package with the official ONDRI colors.

ondridown

This repository houses a lightweight package to provide an RMarkdown template so that we can have harmonized looking documents.

Standards R package

This repository houses an R package that emphasizes structural data checks, primarily formatting, filenames, valid MISSING codes, checks for complete data packages (e.g., DATA, DICT, README). This R package does not perform project-specific checks (e.g., for valid participants or date ranges). Please see the Standards App for (some) of those specifics.

Standards (Shiny) App

This repository houses a heavy-duty app to check data standards for ONDRI and related projects, as well as custom projects and basic data checks.

Data preparation (Shiny) App

This repository houses an app meant primarily for some fundamental but rudimentary data inspection and preparation. This app is meant for use with ONDRI data and, in particular, as the step between the standards checks and the outlier analyses (though this app can be used for inspection and preparation for other analyses).

Outlier analyses (Shiny) App

This repository houses an app meant primarily for implementation of the ONDRI outliers detection pipeline. In short, it provides visualization and report generation based on the techniques found in the Outliers and Robust Structures (OuRS) package.

External materials

Some of the packages and utilities here depend on materials found elsewhere. In most cases these are utilities generally found (e.g., tidyverse). Some specific materials that are necessary for much of the NIBS pipelines are found at Derek’s github page and include:

• The Outliers and Robust Structures (OuRS) package

• The Generalized singular value decomposition package

• The Generalized partial least squares package

And more!

This organization page will also include apps and packages for other tools as they are developed.