This repository hosts resistance databases for several non-tuberculous mycobacterial species as well as kmers which can be used to identify species. The databases are hosted on github to promote collaboration and knowledge sharing.
If you think a mutation/gene should be removed or added please raise and issue here.
Mutations/genes for existing or new can be added by submitting a pull request on a fork. If that previous sentence made no sense to you then you can suggest a change using an issue and we will try help.
With analysis pipelines pretty much standardised, it is evident that accuracy of prediction is affected mostly by the underlying library of mutations. As new evidence for the inclusion or exclusion of mutations is generated, there is a constant need to update and re-evaluate the mutation library. Moreover, it is important for the control of the library to be put in the hands of the end-users. By hosting the library on a separate repository (rather than buried deep in the profiling tool code) it makes it easier to find out exactly which mutations are present. Additionally, github has a number of useful features which can be utilised:
tl;dr - Hosting it here makes it easier to update the library.
It can be used as a direct input to NTM-Profiler, but can also be used for any custom pipeline.
The databases are hosted in db. The ntm_db.kmers.txt file contains the kmers which have been associated with different species. Subfolders, such as Mycobacteroides_abscessus, contain species specific resistance databases (look for the csv files), as well as the reference genome and annotation files as well as some other files which are used by NTM-Profiler.