datasets

A convenient place to park small/test datasets

Naming convention and metadata

Each dataset should adopt this naming convention, composed of underscore-separated tokens and appropriate file extension:

species_samplekeyword1-samplekeyword2-samplekeyword(n)_objectype.ext

Tokens:

1. Species, eg. human, mouse, etc.
2. Sample keyword(s) (separate by hyphens)
3. Type of object/generated by eg. Seurat, DESeq2, Scanpy

Examples:

human_pancreas-islet-T1D_DESeq2.Rds
mouse_brain-cortex-MAD-KO_Seurat.Rds

An associated descriptive markdown file should exist, per dataset. Same file name, but with .md extension.

How to add files to this repo via Git LFS, in GitHub

https://docs.github.com/en/repositories/working-with-files/managing-large-files/configuring-git-large-file-storage

In short:

1. Clone the repo to your local
2. At CLI, type in git lfs track "*.Rds", for Rds files, for example
3. Commit/push the local .gitattributes
4. Add file to repo, git add file.Rds
5. Commit and push