GATK 4 GenomeAnalysisTK still available? (Unable to access jarfile)

Question

[marongiu.luigi]

Dear all,

I am trying to run GATK; I understand the first step is to run a realignment. I removed duplicates with sambamba markdup and I am now on the step for GATK's GenomeAnalysisTK. I first wanted to see if the executable is available thtough GATK so I wrote:

$ java -jar ~/src/gatk/GenomeAnalysisTK.jar
Error: Unable to access jarfile ~/src/gatk/GenomeAnalysisTK.jar

Since the path is given, the following post is not relevant. In this post, I understand that the GenomeAnalysisTK is no longer part of the package and I need to get it independently of GATK.

Is that truly the case or has GenomeAnalysisTK be superseded by another command and, if yes, which one?

Thank you

For completion, the ~/src/gatk directory contains:

$ ls ~/src/gatk/
gatk                           gatk-package-4.0.4.0-local.jar
gatk-completion.sh             gatk-package-4.0.4.0-spark.jar
gatkcondaenv.yml               gatkPythonPackageArchive.zip
GATKConfig.EXAMPLE.properties  README.md

For the version in use:

$ gatk -version
Using GATK jar /home/gigiux/src/gatk/gatk-package-4.0.4.0-local.jar
Running:
    java -Dsamjdk.use_async_io_read_samtools=false -Dsamjdk.use_async_io_write_samtools=true -Dsamjdk.use_async_io_write_tribble=false -Dsamjdk.compression_level=2 -jar /home/gigiux/src/gatk/gatk-package-4.0.4.0-local.jar -version
Exception in thread "main" java.lang.IncompatibleClassChangeError: Inconsistent constant pool data in classfile for class org/broadinstitute/barclay/argparser/CommandLineProgramGroup. Method lambda$static$0(Lorg/broadinstitute/barclay/argparser/CommandLineProgramGroup;Lorg/broadinstitute/barclay/argparser/CommandLineProgramGroup;)I at index 43 is CONSTANT_MethodRef and should be CONSTANT_InterfaceMethodRef
    at org.broadinstitute.barclay.argparser.CommandLineProgramGroup.<clinit>(CommandLineProgramGroup.java:19)
    at org.broadinstitute.hellbender.Main.printUsage(Main.java:380)
    at org.broadinstitute.hellbender.Main.extractCommandLineProgram(Main.java:354)
    at org.broadinstitute.hellbender.Main.setupConfigAndExtractProgram(Main.java:180)
    at org.broadinstitute.hellbender.Main.mainEntry(Main.java:202)
    at org.broadinstitute.hellbender.Main.main(Main.java:289)

[marongiu.luigi]

I guess this post might shed light, is that so? GenomeAnalysisTK has been replaced by RealignerTargetCreator and IndelRealigner?

[Ram]

The jar file is still GenomeAnalysisTK.jar, so no, it has not been "replaced". Please edit your post and mention the GATK version you're using.

[zx8754]

Related / Possible duplicate post: GATK: Unzipped files Cannot Find GenomeAnalysisTK.jar?

[marongiu.luigi]

I have seen that post but I think it relates on how to launch GATK and see the available tools. I would like to know if GenomeAnalysisTK -- which is reported in some manuals and in reference article such as Van Der Auwera et al. Curr Protoc Bioinformatics 2014 -- is still the right tool for the job or I should use these novel ones...

[ATpoint]

What is the output of ls ~/src/gatk/ ?

Answer 1

[prasundutta87]

Please use the latest version of GATK4 for any variant calling job (https://github.com/broadinstitute/gatk/releases).

The variant calling workflow for Germline SNPs+Indels has also changed (https://software.broadinstitute.org/gatk/best-practices/workflow?id=11145)

The quick start guide may be helpful to start testing/running the tools: https://software.broadinstitute.org/gatk/documentation/quickstart

The reference article (Van Der Auwera et al. Curr Protoc Bioinformatics 2014) is old and should be not be followed as the base commands and workflow has been changed a lot. Indel realignment is not necessary now because of Haplotypecaller doing the local reassembly itself. The article can be referred for knowing the reasons behind some steps and get a good idea behind GATK variant calling pipeline, but commands have changed a lot since then.

[marongiu.luigi]

Thank you. So if I understood properly, what was run with

java -jar GenomeAnalysisTK.jar -T <tool> <options>

is now run with

gatk <tool> <options>

is that right? The workflow remains the same.

[prasundutta87]

That is right. Although, the workflow has changed in some places.

The whole indel realignment step has been removed due to the reason I specified above. Also, for multisample variant calling (also called joint variant calling), the GVCF mode is used during the usage of Haplotypecaller. To know about their new pipeline and methods, you can access their workshop slides which are open to all- https://drive.google.com/drive/folders/1lMc4nQIdiuG91GNJiWSdgQFbA4ioZNr6

As mentioned before, the write up under the workflow is important: https://software.broadinstitute.org/gatk/best-practices/workflow?id=11145

To understand how each tool works and their actual command, this page is helpful: https://software.broadinstitute.org/gatk/documentation/tooldocs/current/