Skip to content

Latest commit

 

History

History
100 lines (64 loc) · 4.54 KB

HOWTO.md

File metadata and controls

100 lines (64 loc) · 4.54 KB

Project template

This is a data management templete for github used by the Neuroimaging Statistics Oxford (NISOx) reserach group. This README file should cointain the most relevant information, and also a link to the publication.

Getting started

To get a copy of this template for your own project, please use GitHub import utility available at: https://github.com/new/import, and:

This will ensure that your project is independent from this template repository (rather than being a fork).

Markdown

A Github markdown cheat sheet is available from here.

Folder structure

There should be a general consensus on the folder structure, but extensions and adjustments are are possible. If a dataset that is analyzed is open, it should contain a reference to the source, in the README.md

.
|-- CITATION
|-- README.md
|-- LICENSE
|-- requirements.txt
|-- data
|   -- responde_time.csv
    |   -- raw
    |   -- processed
|-- results
|   -- 10_node_network
    |  -- subj_01_node01.txt
    |  -- subj_01_node02.txt
|   -- summarized_results.csv
|   -- ProjectName.Rmd
|-- figures
    -- fig1.png
    -- fig2.png
    -- time-series_ID057.png
|-- src
|   -- specific_analysis.py

README.md

Mandatory. A description of the repository, its citation, and information to help a user to get started with contents as quickly as possible.

CITATION

Mandatory. Provide information on how users of this shared repository should reference the the work. It may be the bibliographic reference of the paper that this repository supports, and/or data or software descriptor paper for the work.

LICENSE

Mandatory. The lab default license for the data shared is CC-BY, though ensure that the source of the data is compatible with this. The lab default license for any scripts and code is MIT, though pay special care that if code is derived from GPL-licensed code then the license must be GPL.

requirements.txt

Mandatory. List software tools and packages that the repository depends on. If using Python, use pip syntax.

data directory

Mandatory. Original source data and preprocessed data. Summary data and derived data directly used to produces should results directory.

Organisation. Simple projects may not need any subfolders, while large [....]

It is recommended that neuroimaging data should be represented in BIDS format. Consider a "raw" folder that includes unprocessed versions of your data.

results directory

Mandatory. Summary data, derived data, scripts, notebooks, figures and tables. The data in this directory is that needed to produce the results reported in paper text, tables and figures; the notebooks or scripts are those used to create said results.

General functions and scripts used to pre-process and generate derived data should be in the src directory.

Be sure to clearly name tables and figures. Please use the following conventions:

  • Fig_<num>_<label>.<ext>
  • Tab_<num>_<label>.<ext>

For figures the best format is PDF, followed by PNG or TIFF. Avoid JPEG at all costs.

For tables, always provide a CSV or TSV with column headers; if possible use full precision (i.e. avoid saving rounded values that may have been used for typsetting a table for publication).

src directory

Mandatory. Library functions and scripts used to pre-process and generate derived data. Scripts that directly generate results (tables or figures) should be in results.

Notebooks

There are different types of notebooks, and some are directly rendered by Github.

R Notebooks

For R Notebooks, RawGit serves raw html files directly from Github. To get an URL enter the link to the html files created by knit. Here is an example:

  • R Notebook
  • RawGit provides a URL for development, use this so new changes you push will be reflected within minutes.

Further readings