brentp / smoove

structural variant calling and genotyping with existing tools, but, smoothly.
Apache License 2.0
222 stars 21 forks source link

Required tag PRPOS not found #157

Open Saeideh-Ashouri opened 3 years ago

Saeideh-Ashouri commented 3 years ago

Dear @brentp I used the following command to run Smoove on my several thousands of samples: $SMOOVE call --outdir $OUTDIR --exclude $BED --name $ID -f $REF -p 1 --genotype $BAM

but when merging Smoove SV calls of all samples, the program exited with exit status 1 and this error message: Required tag PRPOS not found. Please ensure you've run lumpy withthe -P option to emit breakpoint probabilities.

Is there anything I forgot to do in SV calling step? I hope I won't have to do calling SVs again as the number of samples is quite high. Any help would be highly appreciated.

Best, Saeideh

brentp commented 3 years ago

Hi, can you show the commands you run (just for one sample is fine) and the output you got? This can mean that one or a few samples have truncated files. As a start you could try:

zgrep -cv ^# /path/to/*.persample.vcf.gz

and see if any samples have fewer than expected rows.

Saeideh-Ashouri commented 2 years ago

Dear @brentp

Thank you so much for your prompt reply. The result of zgrep command for most of samples was between 3700-7000 but for 100 samples it was exactly 29,016! So, I repeated the SV calling for those 100 samples again and now the range is 3700-7000 for all of them and merging step is being run. Thank you for your kind help.

Just another question: the merging command is taking a long time (which I expected as the number of samples is quite high [11,000 samples]), it's 25 hours now. I just wanted to know if there is any other way for doing this step so that it takes a shorter time (something like dividing samples into batches etc.).

Best, Saeideh

brentp commented 2 years ago

Hi Saeideh, yes, you can divide into batches to improve the merging time. I would like to improve this at some point, but for now, you'll have to write your own code to do the merging in smaller batches.

I am glad to hear that you got the previous problem sorted.