README

This git repository has been built using the template-research-project template or is the template-research project.

Project organization

The project structure follows the recommendation of quick guide to organize your computation biology project

1. data contains the raw data of your project. If the data is too large, do not commit it (and ideally, create a small script to automatically download the data). If the data is very very large, and you cannot even download it on your computer, try to have a small subsample that you can commit here, and symlink (on the cluster) the large data in this folder. e.g., if you work on the whole set of complete genomes of the NCBI, create a small subsample:

   data/NCBI-partial-dataset

   And add, on the server, a symlink to the whole dataset such that the structure of the folder is:

   data/
    NCBI-partial-dataset
    NCBI-full-dataset
   

   Code such that your scripts work transparently locally on the small subsample data, and remotely on the full dataset (by using snakemake for instance).

2. scripts contains a series of folders, each corresponding to a step in the data analysis process. In those folders, add all scripts necessary to this step. That includes visualization. A concrete example: working on Hi-C data, the first step is always to normalize the data. I thus create a folder normalization in scripts. In it, I add a script called normalize_counts.py and another one called plot_contact_counts.py. The first is the script I use to normalize the data. The second is a script I use to visualize the raw or the normalize contact counts: it outputs a png file for each of my datasets which I can then add to my labnotebook (more below).

3. results contains the labnotebook. This notebook should either by an html file or something that can be compiled in an HTML file so that it is easily shareable with collaborators.o

4. doc contains any communication document you may be creating relating to this project.

Check out the original publication to understand the additional folder

The electronic labnotebook

Keep track of all and any experiment your run, whether it fails or not! Do this in an electronic labnotebook, so that you can easily share it with collaborators. This is particularly useful if you are working remotely and/or have video confcall meeting regurlarly.

This labnotebook will be built automatically, and the resulting HTML will be pushed on the gh-pages.

• This repository's labnotebook is done using sphinx, but anything that builds in HTML in results/_build/html using make all will work with the current github workflow setup.
• Try to write something everyday.