jodyphelan / TBProfiler

Profiling tool for Mycobacterium tuberculosis to detect ressistance and strain type from WGS data
GNU General Public License v3.0
104 stars 43 forks source link

inconsistency between Resistance report and Resistance variants report #375

Open alantsangmb opened 3 months ago

alantsangmb commented 3 months ago

Hi @jodyphelan I noticed some inconsistency between the Resistance report and the Resistance variants report. Specifically, for my attached result generated from v6.2.1, the pncA p.Val21Ala is listed as "Not assoc w R - Interim" in the Resistance variants report, but it is listed as Resistance in the resistance report

I have cross-checked the WHO catalogue and confirmed that this mutation is indeed "Not assoc w R - Interim" to Pyrazinamide.

Sample3576.json Sample3576.results.zip

jodyphelan commented 2 months ago

It seems related to the fact that the default tbdb library comes from two sources (more on that here). It seem to be designated as not associated via the WHO catalogue but was previously reported in the literature as associated with resistance. Looking at the library there seem to be three of these cases:

Gene Mutation type drug original_mutation confidence source comment
pncA p.Val21Ala drug_resistance pyrazinamide p.Val21Ala Not assoc w R - Interim tbdb Mutation from literature
pncA p.Thr160Ala drug_resistance pyrazinamide p.Thr160Ala Not assoc w R - Interim tbdb Mutation from literature
pncA p.His137Arg drug_resistance pyrazinamide p.His137Arg Not assoc w R - Interim tbdb Mutation from literature

Some of these could potentially be conferring resistance, but perhaps the best solution here would be to remove these from the resistance list.

alantsangmb commented 2 months ago

Thank you so much. What is the best way to remove these items from the resistance list?