DepthOfCoverage and DiagnoseTargets: our plans for development and support

We have decided to continue providing and supporting DepthOfCoverage and DiagnoseTargets for the foreseeable future. Going forward, we'll try to integrate them and develop their features to address the main needs of the community. To this end we welcome your continuing feedback, so please feel free to contribute comments and ideas in this thread.

To all who took the time to tell us what you find useful about DoC and DT (and what you wish it could do), a big thank you! This is always very useful to us because it helps us identify which features are most valuable to our users.



mengyuankan


Thanks Geraldine. A newbie question: what's the difference between DepthOfCoverage and DiagnoseTargets, and which one do you recommend to use (or both)?

Tue 12 Mar 2013

Geraldine_VdAuwera


The two are somewhat convergent, but basically DepthOfCoverage allows you to evaluate the depth of coverage in your data at certain sites or over intervals generally, whereas DiagnoseTargets is more specifically designed to evaluate the quality of data covering exome target intervals. The best thing to do is to read their respective documentations pages (links below) and see which sounds like it will suit you needs best. http://www.broadinstitute.org/gatk/gatkdocs/org_broadinstitute_sting_gatk_walkers_coverage_DepthOfCoverage.html http://www.broadinstitute.org/gatk/gatkdocs/org_broadinstitute_sting_gatk_walkers_diagnostics_targets_DiagnoseTargets.html

Tue 12 Mar 2013

kevyin


Hi @Geraldine_VdAuwera and thanks for support on this. The DepthOfCoverage documentation mentions -nt as a parallelism option but I get an error. `##### ERROR MESSAGE: Invalid command line: Argument nt has a bad value: The analysis DepthOfCoverage aggregates results by interval. Due to a current limitation of the GATK, analyses of this type do not currently support parallel execution. Please run your analysis without the -nt option.` My command line: `java -jar GenomeAnalysisTK.jar -T DepthOfCoverage -I ExampleBAM.bam -R exampleFASTA.fasta -o ./test_coverage_out -nt 4` Is this an error in the documentation? Or am I missing something? Thanks.

Tue 12 Mar 2013

Geraldine_VdAuwera


Hi @kevyin, DOC is currently set up to aggregate statistics over intervals by default, which is incompatible with the `-nt` mode. You can disable this behavior by using the `--omitIntervalStatistics` flag, which should make `-nt` work. Let me know if you have any issues with that. We may change the default behavior in future; in any case we will add a note about this to the documentation. Thanks for reporting this!

Tue 12 Mar 2013

SerenaRhie


Hello, I want the concise depth of coverage summary as well as the per-location covering number of bases for my RNA-Seq data. With not splitting the N containing reads, what does DoC exactly do with -U ALLOW_N_CIGAR_READS option? Is it still counting the bases in the CIGAR array represented as matched? I think the bases represented as N in CIGAR should not be counted, just as the deletions (D) are treated. (Maybe an option such as --includeNBases might be appropriate for special cases.) A good explanation will be helpful to decide the DoC or not. And please do something with the last line of the ERROR MESSAGE "Notice however that if you were to choose the latter, an unspecified subset of the analytical outputs of an unspecified subset of the tools will become unpredictable.", which I think is totally nonsense! :smile:

Tue 12 Mar 2013

Geraldine_VdAuwera


@SerenaRhie, Hah, I think the developer who wrote that error message was feeling particularly sarcastic that day. I'll see what we can do to make that a little more informative. Regarding the behavior question, I believe the Ns in the CIGAR are correctly interpreted as not contributing any coverage. That said, you could run DoC after running the SplitNTrim step, which will give you effective depth after processing.

Tue 12 Mar 2013

SerenaRhie


@Geraldine_VdAuwera Thanks!

Tue 12 Mar 2013



Search blog by tag

2013 ad agbt14 appistry ashg ashg2014 baserecalibrator belgium best-practices beta blog brussels bug bug-fixed cancer catvariants challenge combinegvcfs combinevariants commandline commandlinegatk commercial compbio competition conferences denovo depthofcoverage diagnosetargets downtime error fastaalternatereferencemaker fix forum gatk gatk-3-0 gatk-3-2 gatk3 genotype genotypegvcfs genotyperefinement gsa gvcf haploid haplotypecaller hardware holiday htsjdk ibm indelrealigner job job-offer jobs joint-analysis joint-discovery key license media meetings mendelianviolations multisample multithreading mutect nt pairhmm paper performance phone-home picard pipeline ploidy polyploid poster presentations press printreads queue randomlysplitvariants readbackedphasing reducereads reference-model release release-notes rnaseq search selectvariants service slides snow spam speed splitncigarreads status sting support syntax talks team third-party-tools topstory trivia troll unifiedgenotyper userstories validatevariants variantannotator variantrecalibrator variantstobinaryped version-highlights versions video videos vqsr webinar workshop


GATK Dev Team

@gatk_dev

@bbongalov Thanks, glad you're finding it useful :)
29 May 15
@ewanbirney @BioMickWatson @luispedrocoelho @MattBashton GATK 3.4 has CRAM r/w, no indexing (so: meh). GATK4 will have full CRAM thru htsjdk
29 May 15
@bbongalov True, #GATK is a bit of a control freak. But it's only because it cares so deeply about doing the right thing.
28 May 15
@torstenseemann @yokofakun Is this a raw call or after filtering? Would have to see annotations to opine.
28 May 15
Holiday notice -- #GATK forum on hiatus for Memorial Day weekend, service resumes May 26. https://t.co/uVpA8SSXdl
23 May 15