bcgsc / HLAminer

⛏ HLA predictions from NGS shotgun data
Other
51 stars 15 forks source link

Alleles not produced in HLAminer output #32

Closed Addicted-to-coding closed 5 months ago

Addicted-to-coding commented 8 months ago

Hi, I am trying to run HLAminer on a set of .fq files. I first index the files using bwa to produce a .sam file. Then, I run HLAminer according to the following command: bin/HLAminer.pl -h database/HLA-I_II_GEN.fasta -s 500 -q 1 -i 1 -p database/hla_nom_p.txt -a ERRfile.sam

This produces an output as follows

Screen Shot 2024-02-14 at 2 01 57 PM

How can I produce an output similar to the one displayed in the documentation, with the alleles listed?

warrenlr commented 8 months ago

thank you for your question.

Did you run the demo? If not, I strongly recommend that you start there (after reading all documentation) and compare your results with those in the test-demo folder.

Once you confirm that the demo runs, you should run HLAminer by replacing the fq with your files.

Empty predictions are often a symptom of either 1) errors in execution of the pipeline, 2) inadequate data or 3) too low depth of coverage.

warrenlr commented 5 months ago

Closing due to inactivity (>3Mo).

It is important to generate an MD tag in the alignment output. The tool requires it for predictions. Take a look at the demos for the bwa and minimap2 commands. Alternatively, the score could be lowered.

e.g. minimap2 -t 60 -ax map-ont --MD