search

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

Prediction issue #346

Closed RehmanZaira closed 4 months ago

RehmanZaira commented 5 months ago

I have installed TBProfiler locally and run number of samples. All the samples turned out to be sensitive. When the same samples run on web version of TBProfiler the same samples reported to be either MDR or XDR. Which is actual results as the samples are in fact MDR and XDR proven through DST method. Why this discrepancy exist? How can I improve the results of locally installed TBProfiler?

jodyphelan commented 5 months ago

Hi @RehmanZaira ,

Could you let me know the command you are using?

RehmanZaira commented 5 months ago

Hi @RehmanZaira ,

Could you let me know the command you are using?

I am processing the nanopore sequencing reads from a directory and here is the command:

"$TB_PROFILER" profile -1 "$file" -p "$PREFIX"_"$sample_name"
jodyphelan commented 5 months ago

Try adding --platform nanopore and see if that fixes the issue?

RehmanZaira commented 5 months ago

Now it is working fine. Thank you for the help. I have another query: Can we change the threshold of variant calling in pipeline?

jodyphelan commented 4 months ago

Yes you should be able to change this. Check out https://jodyphelan.github.io/tb-profiler-docs/en/algorithm/#variant-calling

RehmanZaira commented 4 months ago

Thank you so much for the help.