RASE DB skeleton

This repository contains the RASE DB skeleton for creating new databases for rapid inference of antibiotic resistance and susceptibility using RASE. Other components of the software include the RASE pipeline and the core RASE package. For more information, see the RASE paper and the RASE supplementary materials.

Creating a custom RASE DB

1. Set up your computational environment.

2. Create a repository for your DB on some git-based repository hosting service such as Github, Gitlab, or Bitbucket.

3. Run the following script in an empty directory. It will create a local git repository with with the RASE DB skeleton.

   /usr/bin/env bash -c "$(curl -fsSL https://raw.githubusercontent.com/c2-d2/rase-db-skeleton/master/init-repo.sh)"
   

4. Interconnect the local repository with the remote one. For Github, this could be done by

   git remote add origin [email protected]:your-github-id/your-db-repo.git
   git push -u origin master
   

5. Test that your computational environment is configured properly by constructing an example database:

   make && make cleanall
   

6. Modify the main configuration file conf.mk. In particular, update the DB name, the list of antibiotics and their corresponding breakpoints.

7. Place your input data or code for their download into published/. If some of the data are private, copy them to unpublished/.

8. Adjust code for preparing isolate whole-genome sequences in isolates/. Test it by running make inside the directory.

9. Adjust code for preparing the phylogenetic tree in tree/. Test it by running make inside the directory.

10. Adjust code for preparing the resistance data in resistance/. Test it by running make (or make step1, make step2, etc. for individual steps) inside the directory.

11. Construct the database by running make in the root directory. The constructed database then can be found in _output.

12. You can optionally plot MICs and resistance categories on a tree, and calculate prevalence of resistance phenotypes by make plots.

Now your code for building the database is ready. All the outputs can be removed by make clean or make cleanall (removes even downloaded source files), and regenerated by running make in the root directory.

FAQs

I have only S/I/R information rather than full MICs. How should I proceed?

Use fake MICs. For instance, you can use 0.0 for susceptible isolates, and 999.0 for resistant and intermediate ones.