We've moved!
This site is now read-only. You can find our new documentation site and support forum for posting questions here.
Be sure to read our welcome blog!

FastaAlternateReferenceMaker WARNINGS


first of all thanks for the support offered here. But (sadly) I have an other two questions.

I used the FastaAlternateReferenceMaker to create a consesus sequence and it worked great. Here I have a question regarding a warning: "IndexDictionaryUtils - Track variant doesn't have a sequence dictionary built in, skipping dictionary validation". I do not get this warning. Which dictionary is needed here. Is this warning comming up, because I used .gz and .tbi compressed vcf-files?

And a furhter question regarding the creation of a consesus sequence. I`m unsure what to do in cases of heterozygous INDELs. What would you suggest to do? Skip them as they are only represented by one allele or include them in the consensus. Both seems to be incorrect. Do you know the general handling here?


Best Answer


  • tommycarstensentommycarstensen United KingdomMember ✭✭✭

    I get the same warning some 30k times when running VariantRecalibrator3.4. I'm assuming it's safe to ignore.

  • tommycarstensentommycarstensen United KingdomMember ✭✭✭

    Please ignore my question. I already asked about it here. Previously VR3.3 took me ~12 hours to run with -nt 24. Now it will take me several weeks according to the stderr and it has already run for more than 8 hours. I must somehow have done something wrong or there is some serious I/O load on our disks right now. I will try out VR3.3.

  • tommycarstensentommycarstensen United KingdomMember ✭✭✭

    Sorry about the spam above. I solved my problem. I reduced the number of VR input files from ~30k to ~3k by concatenation (with a competing tool) and the runtime dropped to hours instead of weeks.

Sign In or Register to comment.