It looks like you're new here. If you want to get involved, click one of these buttons!
skblazer
Posts: 18Member ✭
I tried to run several HaplotypeCaller jobs simultaneously on a node. But sometimes, not all the time, they caused the nodes collapse. I noticed that HaplotypeCaller sometimes occupied tens of GB of memory, is this normal? The average depth for my samples is 30x-50x.
ebanks
Posts: 497 mod
Currently the Haplotype Caller does not do any down-sampling (although it's something we are planning to add in the future). So while the average coverage is 30x, there will certainly be regions of the genome where coverage spikes significantly and can require a lot of memory to run.
