Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Rc vs 1365 hatcher testing parquet #8845

Draft
wants to merge 42 commits into
base: ah_var_store
Choose a base branch
from

Conversation

RoriCremer
Copy link
Contributor

@RoriCremer RoriCremer commented May 23, 2024

TODO:
Builder method
Still want to fully test the compressed location method (do we? do we care about this for Azure?!??!)
Still want to fully test the header creation

to run this locally:
java -jar ../build/libs/gatk.jar
CreateVariantIngestFiles
-V HG00405.haplotypeCalls.er.raw.vcf.gz.rb.g.vcf.gz
-L wgs_calling_regions.hg38.noCentromeres.noTelomeres.interval_list
-IG FORTY
--force-loading-from-non-allele-specific false
--ignore-above-gq-threshold false
--project-id gvs-internal
--dataset-name hatcher_import_test
--enable-reference-ranges true
--enable-vet true
--enable-vcf-headers true
--output-type PARQUET
-SN SM-N58BO
-SNM sample_map.csv
--ref-version 38
--skip-loading-vqsr-fields false

jamesemery and others added 30 commits October 11, 2023 10:01
* hmer ondel must have mon length

* Revert "hmer ondel must have mon length"

This reverts commit 7852871.

* remove superfluous variant type condition

* fix error message to actually reflect missing argument

* fixed unittest to include variant type

* Remove conflict
* Additional fix + logging fixes
* Added missing initialization
)

* Add option for keeping disjoint mates in ASM

* Better name and fixing reports

* Finish fixing report

* Fix report name
New Tool: GroundTruthScorer
Update: FlowFeatureMapper
…dependency on the ADAM library (#8606)

* Add a native GATK implementation for 2bit references, with comprehensive unit tests

* For now, this is only hooked up to the Spark codepath, but it could easily be hooked up to ReferenceDataSource and the Walker codepath as well

* Remove the dependency on the ADAM library, to resolve conflicts with future dependency upgrades
…curity scanner to build.gradle (#8607)

* Updated many GATK dependencies to address known security vulnerabilities

* Added a security scanner to build.gradle

* There are still some remaining vulnerabilities in GATK dependencies, but this eliminates most of them
* Update http-nio and wire it so it's configured at startup along with GCS setttings.
* New experimental tool to print out human readable file diagnostics for cram/crai/bai files.
…#8438)

* GATK's lack of support for az:// URIs means that although GenomicsDB can
  natively read them, parts of the java code crash when interacting with them
* Adding --avoid-nio and --header arguments
  These allow disabling all of the java interaction with the az:// links
  and simply passing  them through to genomicsdb
  This disables some safeguards but allows operating on files in azur
* Update GenomicsDB version to 1.5.1 for azure improved support

* There are no direct tests on azure since we do not yet have any infrastructure
  to generate the necessary tokens, there is a disabled test which requires
  #8612 before we can enable it.

---------

Co-authored-by: Nalini Ganapati <[email protected]>
Co-authored-by: Nalini Ganapati <[email protected]>
For having variable ploidy in different regions, like making haploid calls outside the PAR on chrX or chrY, 
there is now a --ploidy-regions flag. The -ploidy flag sets the default ploidy to use everywhere, and --ploidy-regions
should be a .bed or .interval_list with "name" column containing the desired ploidy to use in that region
when genotyping. Note that variants near the boundary may not have the matching ploidy since the ploidy used will be determined using the following precedence:

* ploidy given in --ploidy-regions for all intervals overlapping the active region when calling your variant
  with ties broken by using largest ploidy); note ploidy interval may only overlap the active region and determine 
  the ploidy of your variant even if the end coordinate written for your variant lies outside the given region
* ploidy given via global -ploidy flag
* ploidy determined by the default global built-in constant for humans (2).

---------

Co-authored-by: Ty Kay <[email protected]>
Co-authored-by: rickymagner <[email protected]>
* Update the GATK base image to the latest Ubuntu LTS release (22.04)

* Add some additional useful utilities to the base image

* Switch to a newer conda version with a much faster solver

* Update the scripts and documentation for building the base image

* Update the VETS integration tests to allow for a small epsilon during numeric comparisons, and include the full diff output in exception messages when a mismatch is detected
…oud-based docker build, and add a release script (#8247)

* Added a -r argument to build_docker_remote.sh to toggle the RELEASE flag during
  docker builds

* Added a release_prebuilt_docker_image.sh to release a prebuilt docker image to the
  official repos
* update to htsjdk 4.1.0 which enables http-nio in more cases
* remove several test cases handling genomicsdb path parsing which were testing nonsensical paths that are now illegal
* This should make http access seamless in many places

* The way this handles query parameters is not ideal for signed url cases so we'll need to revisit that
…ervals output (#8621)

* Write gCNV interval output ID=GT header as Type=String

Incorrectly writing this as Type=Integer causes bcftools to misparse
the genotype field.

* Use correct header types and numbers in test VCF file
* include normal seq error log likelihood in Permutect dataset

* handle different alelle representations in multiallelic / indel variants for Permutect training data mode

* set the default artifact to non-artifact ratio to 1 in Permutect training data mode
@RoriCremer RoriCremer force-pushed the rc-vs-1365-hatcher_testing_parquet branch 3 times, most recently from d8c18a6 to ddd1141 Compare June 10, 2024 20:41
@RoriCremer RoriCremer force-pushed the rc-vs-1365-hatcher_testing_parquet branch 5 times, most recently from de9c85c to edd6938 Compare June 17, 2024 19:28
@RoriCremer RoriCremer force-pushed the rc-vs-1365-hatcher_testing_parquet branch from 4569795 to e852dcf Compare June 26, 2024 01:11
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.