Tips for project management in bioinformatics
2 min readJun 21, 2022
When you start a new project, you will probably make a custom folder structure with a high probability of getting lost in a couple of months.
Here I give some tips to avoid making too much mess.
First of all, you will need to make a unique folder name for your project. A combination of time-stamp and a meaningful name is sufficient.
Then you can place this in a folder structure for being more organized, I choose this: group_name/user_name/project_name
Each time you have a new project you can use a simple folder structure for keeping separate different entities:
- Input data. Never touch them, they should be ideally the same as the ones you receive. Check them with md5sum to be sure the copy is ok.
- Source and programs. Here you can store the tools used for the project. Virtual environments, containers, scripts, executables, etc... You can have a physical copy or in case of big and immutable files, you can have soft links.
- Docs. Here you can have a README file in which you write down some text about this specific project, you can have a local copy of the protocol used, the interesting papers, etc. This is the space for talking to a “future you”.
- Analysis. Here you have all the attempts, tests, and intermediate files generated for the analysis. You can also add some files explaining a bit the command lines used with some descriptions. You can have soft links to the Source folder to single sub-folders for connecting the executables with the particular analysis.
- Results. Here are the final results, I suggest you make soft links to files inside the Analysis folder or a copy if you need to remove that folder. This is what you want to communicate to other people what is the final result of your project. You can also link the README file from the docs and publish this folder on a web server for delivering the results.
I made a simple shell script for generating this simple folder structure here:
