Awesome Open Source
Awesome Open Source

User Support:

For user questions please look for answers and ask first in the GATK forum.

Coverage Status Build Status License: MIT

A set of Java command line tools for manipulating high-throughput sequencing (HTS) data and formats.

Picard is implemented using the HTSJDK Java library HTSJDK to support accessing file formats that are commonly used for high-throughput sequencing data such as SAM and VCF.

Picard now builds and passes tests under Java 11. This should be considered to be a Beta feature. As of version 2.0.1 (Nov. 2015) Picard requires Java 1.8 (jdk8u66). The last version to support Java 1.7 was release 1.141.

Building Picard

  • First, clone the repo:
    git clone
    cd picard/
  • Picard is now built using gradle. A wrapper script (gradlew) is included which will download the appropriate version of gradle on the first invocation.

  • To build a fully-packaged, runnable Picard jar with all dependencies included, run:

    ./gradlew shadowJar
  • The resulting jar will be in build/libs. To run it, the command is:
    java -jar build/libs/picard.jar
    java -jar build/libs/picard-<VERSION>-all.jar 
  • To build a jar containing only Picard classes (without its dependencies), run:
    ./gradlew jar
  • To clean the build directory, run:
    ./gradlew clean

Running Tests

  • To run all tests, the command is:
    ./gradlew test
  • To run a specific test, the command is:
    ./gradlew legacyTest --tests "*TestClassName*"
    ./gradlew barclayTest --tests "*TestClassName*"

Running legacyTest uses the legacy commandline parser while barclayTest uses the new parser.

Changing the released version of HTSJDK that Picard depends on

To switch Picard's HTSJDK dependency to a different released version:

  • Open build.gradle
  • Edit VERSION in the following line to be a different released version of HTSJDK. HTSJDK releases are listed here
    final htsjdkVersion = System.getProperty('htsjdk.version', 'VERSION')`
  • Open a pull request with this change

Building Picard with a Custom Version of HTSJDK

During development in Picard, it is sometimes necessary to build locally against an unreleased version or branch of HTSJDK.

  • To build against an unreleased version of HTSJDK's master branch:

    • Go to the Broad artifactory, where continuous snapshots of HTSJDK's master branch are published, and select the version you want to use. For example, 2.5.1-9-g5740ca1-SNAPSHOT. You can search by tag or short git commit hash.
    • In your Picard clone, run ./gradlew shadowJar -Dhtsjdk.version=VERSION, where VERSION is the version of the HTSJDK master branch snapshot you want to use.
  • To build against a version of HTSJDK that has not yet been merged into HTSJDK's master branch:

    • Clone HTSJDK, and in your clone check out the tag or branch you want to build Picard with.
    • Run ./gradlew install printVersion in your htsjdk clone to install that version to your local maven repository. Take note of the version number that gets printed at the end.
    • Switch back to your Picard clone, and run ./gradlew shadowJar -Dhtsjdk.version=VERSION, where VERSION is the version of HTSJDK you installed to your local maven repository.

Releasing Picard

Full instructions on how to create a new release of Picard are here

Path providers

Picard has limited support for reading from Path providers. Currently only google's api is supported, and only a few tools support this. To run with this support you need to compile the cloudJar target with gradle:

./gradlew cloudJar

then run picard as follows:

java -jar build/lib/picardcloud.jar <Picard arguments starting from program>

For example:

java -jar build/lib/picardcloud.jar CrosscheckFingerprints \
   I=gs://sample1.vcf \
   I=gs://sample2.vcf \
   H=Haplotype_db.txt \

Alternatively, you can run the tool via the GATK which bundles the Google-Cloud jar, and should thus "Just Work".


Please cite this repository when using Picard tools for your publications.

“Picard Toolkit.” 2019. Broad Institute, GitHub Repository.; Broad Institute

  title = {Picard toolkit},
  year = {2019},
  publisher = {Broad Institute},
  journal = {Broad Institute, GitHub repository},
  howpublished = {\url{}}

Identifiers from software registries are increasingly accepted by journals, as in (biotools:picard_tools) or (RRID:SCR_006525).

Picard is migrating to semantic versioning. We will eventually adhere to it strictly and bump our major version whenever there are breaking changes to our API, but until we more clearly define what constitutes our official API, clients should assume that every release potentially contains at least minor changes to public methods.

Please see the Picard Documentation for more information.

Alternative Project Comparisons
Related Awesome Lists
Top Programming Languages
Top Projects

Get A Weekly Email With Trending Projects For These Topics
No Spam. Unsubscribe easily at any time.
Java (391,896
Version (20,615
Sequencing (2,263
Vcf (1,350