Generated html file empty

[dBenedek]

Hello,

I have done the following steps:

1. extract sites:

   for f in "$@"; do
   LOC="/disk/work/shared/data/MDS/${f}/alignment/STAR"
   cd ${LOC}
   for bam in *_Aligned.out_sorted_rg.bam; do
   base=$(basename $bam .bam);
   ${SOMALIER} extract -d ${OUT} --sites sites.hg38.rna.vcf.gz -f ${REF} ${LOC}/${bam}; 
   mv ${OUT}${f}.somalier ${OUT}${f}_${base}.somalier; 
   done
   done

I renamed each file inside the for loop, because each dataset's bam files have the same read group, so the file names would be the same in case of the same dataset.

2. run somalier relate:

   {SOMALIER} relate -g sample_groups.csv /data/somalier/*.somalier

After these steps, there's a .html output file (8.6 MB), however, it is blank, changing axes dows not affect anything.

What would cause this problem?

Thanks in advance, Benedek Dankó

[brentp]

can you share the html file and the tsv output?

you could also try using the --sample-prefix argument to somalier extract so that each somalier file has a unique prefix.

[dBenedek]

Even if I use --sample-prefix=$base the generated .somalier files have the name from the bam files' read group tag.

Output files: somalier.samples.tsv.gz somalier.html.gz

[brentp]

yes, you would change --sample-prefix to be unique for each sample, not to a shared $base.

[dBenedek]

yes, you would change --sample-prefix to be unique for each sample, not to a shared $base.

OK, I will try that, thank you!

[brentp]

ok. please let me know if that solves the problem as that's why --sample-prefix exists.

[dBenedek]

Now the generated .tsv files look OK, but the html file is still blank.

Output files: somalier.samples.tsv.gz somalier.html.gz

[brentp]

ok. I'll make a fix for this for next release. you can edit somalier.html: find Githubissues.

Githubissues is a development platform for aggregating issues.