Heads up:
We’re moving the GATK website, docs and forum to a new platform. Read the full story and breakdown of key changes on this blog.

bruce01 ✭✭

About

Username
bruce01
Joined
Visits
166
Last Active
Roles
Member
Points
54
Badges
10
Full Name
bruce moran

Comments

  • The current documentation (4.1.1.0) specifies the old method including ':'. Doc for 4.0.5.0 shows correct method. On 4.1.0.0 supplied in Docker the new method is correct.
  • Is it possible to have a 'dummy' output? So for example if I want to write a precursor workflow that downloads packages and dependencies, then installs them. In that case, the output might be ${completed_download}.txt, and that is input for the next…
    in output Comment by bruce01 July 2017
  • Or just use --disableSequenceDictionaryValidation true ...
  • Hi, skipped testing other tools because as you note, it is looking for a sequence dictionary, so I used samtools reheader with the genome.dict that I am aligning against and it works. But this is not in the tutorial 6484 files, nor is it specified. …
  • After testing, I get this error: A USER ERROR has occurred: Input files reference and reads have incompatible contigs: No overlapping contigs found. reference contigs = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20…
  • OK, I see some mild discontent from comments on that page. FWIW, and in such a (necessarily) pedantic field, using a file format that is already well known, and the acronym of which actually stands for the opposite of what you are storing in it, see…
  • I had a few further issues with using the same hapmap file as @KostasMavrommatis (thanks for the advice though!) First the -R flag in Tabix is -B (for BED file input). But using this then gave the error: The provided VCF file is malformed at appr…
  • Hi Sheila, just to clarify I was saying I should add DiagnoseTargets to my own pipeline, not that you should add to best practices, sorry for any confusion. Bruce.
  • Hi, sorry, had the question as draft and must have clicked post instead of cancel! I have indeed just used DiagnoseTargets walker and called out all my primary alignments by flag to a new BAM, so best to just use those for coverage stats. Will be in…
  • Thanks @Geraldine_VdAuwera, I was thinking similarly myself about the training set. I find it better to remove spurious variants earlier (for a conservative approach, probably have to do it with relaxed parameters too), my MDS still look good from e…
  • Hi @ami, thanks for your input. As to the RNAseq data, should I then make a separate training set for VQSR? I have essentially 6 replicates per animal (two tissues at three timepoints) so there is pretty good concordance when I do the LD pruning bet…
  • Hi Geraldine, yeah, the MDS plots are basically a way to check that the SNPs in the LD-pruned set are representative of the 'population', or my tiny part of it. The LD-pruning was to try and reduce down to '10s of thousands' of SNPs which I had rea…
  • I was using R_3.x and got errors based on some ggplot2 deprecated functions ('opts' is now 'theme' etc). Plots can be generated by running: sed 's/opts/theme/g' r_script.out.R | sed 's/theme_rect/element_rect/g' | sed 's/theme_line/element_line/g' …
  • Have been testing this all morning and have it working now, unsure of cause of issue. Think it could be a bad *.idx made previously that then caused the error above. Also possibly due to not having full path names in *.list. Once I started from sc…
  • Just as a follow-up I had an errant contig in my reference fasta not in my VCF, so the error was telling me that.
  • Hi Eric, undoubtedly it is an error on my end having had to make my own VCF file following the failure of a BED file previously used successfully on another version of your softwares. I will look at your bundles VCFs and determine what might be my …
  • It is now doing the same thing from my homemade VCF (parsed from ds_flat files). I used vcf-concat to concatenate sorted per-chromosome VCFs. I index with IGVtools but I then get the output INFO 13:45:33,064 HelpFormatter - ----------------------…
  • Hi Eric, the start log: INFO 07:49:19,854 HelpFormatter - -------------------------------------------------------------------------------- INFO 07:49:19,857 HelpFormatter - The Genome Analysis Toolkit (GATK) v2.1-8-g5efb575, Compiled 2012/08/30…
  • Hi Mark, it is literally as I have it above: .#####ERROR------------------------------ and then it quits to cmd line. Bruce.
  • In the example of LB and PI (section 9) should the PI not be PI:200, PI:400, PI:200, PI:400 vs 200,200,400,400? LB flags the library, and you have the 2 libraries sequenced twice, once at 200bp inserts, once at 400bp. Surely this example defines it …