Skip to content

Latest commit

 

History

History
311 lines (240 loc) · 12.7 KB

CHANGELOG.md

File metadata and controls

311 lines (240 loc) · 12.7 KB

Change Log

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog and this project adheres to Semantic Versioning (SemVer). The meaning of each entry is user-based and can be found in our Wiki Versioning page.

Unreleased

1.5.2

Fixed

  • Allow special characters in advance java-property for barcode sequence/quality separators

Developer

  • Remove unused imports

1.5.1

Fixed

  • Add message for unknown command referring to a legacy pre-release tool
  • Documentation update of SRA Hadbook link (previous is currently unavailable)
  • Stacktrace is printed for non-user exceptions to help developers
  • Casava formatted files with non-standard barcode sequences (non-ACTGN) are now supported
  • Better error message for malformed barcode file with multiple-tabs separating header lines
  • Fix typos in documentation

Developer

  • Upgrade gradle to 4.9 and jacoco plugin to 0.8.1
  • Major code cleanup and simplification
  • Rename of RTFastqContstants to RTFastqConstants
  • Rename of ApplyTrimResultReadTransfomer to ApplyTrimResultReadTransformer

1.5.0

Fixed

  • Improved performance for quality checking in every tool
  • Improved performance for trimming pipelines
  • Support non-local reference files for CRAM (e.g., HDFS or GCS)
  • Fix typo in README (building section)

Added

  • Experimental tool to parse GEM-mappability (GemMappabilityToBed)

Developer

  • Upgrade gradle and plugin versions
  • Simplify RTBaseTest and extend functionality with GATK's BaseTest
  • Force locale to US-EN in tests
  • Update HTSJDK to 2.16.0
  • Update to released GATK 4.0.6.0 (patched-version for ReadTools)

1.4.1

Fixed

  • Support non-local FASTQ inputs (e.g., HDFS or GCS)
  • DownloadDistmapResult checks for successful job (_SUCCESS file)

1.4.0

Changed

  • Read filters does not support short arguments anymore (GATK update)

Fixed

  • Improve command line startup information
  • Better error message for invalid tool names

Added

  • Experimental tool for compute mapped pair-end statistics per window (ComputeProperStatByWindow)
  • Add support for long reference sequences in BAM (HTSJDK update)
  • Add support for long CIGAR in BAM (HTSJDK update)
  • Improvement in Intel compression (GKL update)

Developer

  • New framework for read statistics
  • New program group for mapped reads
  • New walker class for reads (SAM/BAM/CRAM) - use only for mapped reads
  • New templates for documentation to integrate experimental/beta features
  • Update HTSJDK to 2.15.0
  • Update to released GATK 4.0.2.1 (patched-version for ReadTools)
  • Update to barclay 2.0.0 and other transitive dependencies from GATK
  • Configuration from GATK is made by the default Owner config file

1.3.0

Fixed

  • Fix failing uncompression of the jar file on OSX due to name collisions
  • More concise repository README without redundant information
  • Clarify usage of DownloadDistmapResult --noRemoveTaskProgramGroup

Added

  • Include validation stringency argument in QualityEncodingDetector
  • Support for Hadoop compression service in the classpath and documentation in help pages

Security

  • Update documentation page dependencies, which had security issues

Developer

  • Exluce junit from gradle configuration
  • Add Wiki for developers (repository level)
  • Add contribution information and code of conduct
  • Remove warnings in help pages (jekyll build)

1.2.1

Fixed

  • Bug fix: remove PG tag in records while DownloadDistmapResult drops PG headers

1.2.0

Fixed

  • User error error for corrupted BAM file instead of unexpected error

Added

  • Trimming support for ReadsToDistmap
  • Support of input CRAM files with reference in HDFS (might fail if used also for output)

Changed

  • Argument expansion for collections is now .args instead of .list
  • Support for several libraries in LibraryReadFilter (GATK v4 change)
  • Update documentation

Developer

  • Refactored trimming engine for improved re-usability
  • Updated jacoco version to 0.8.0
  • Removed VersionEye due to end of the service
  • Using our own classes for argument definitions instead of relying in GATK
  • More test and consistency on them
  • Fix FreeMarker template to properly format jekyll tables

1.1.0 - 2017-10-04

Fixed

  • Fix bug for FASTQ read names with spaces without barcode/pair-end information
  • Honor asynchronous writing in Distmap tools
  • Fix error for different number of barcodes in dictionary and files
  • Major bug fixes to CRAM format (HTSJDK change)

Added

  • Add walkthrough data for documentation
  • Use pre-sorted output for DistmapPartDownloader if possible (speed-up)
  • Better support of Snappy compression/decompression in more systems (HTSJDK change)

Changed

  • Assume SO:unsorted and GO:query for pair-end data if no header is present
  • Do not allow SO:coordinate for pair-end data
  • Do not change SO:queryname for SO:unsorted for single-end data
  • Update documentation
  • Remove PrimaryAlignmentReadFilter in favor of PrimaryLineReadFilter (GATK v4 change)
  • More consistent GCS support (GATK v4 change)

Developer

  • Force HTSJDK dependency to be the one provided
  • Implement automatic documentation framework
  • Remove FastqToReadIterator in favor of generic RecordToReadIterator
  • Add tests for Distmap tools for HDFS
  • Force testing dependencies to be in hte version provided
  • Use walkthrough data for some tests
  • Test clean-up

1.0.0 - 2017-05-18

Fixed

  • MD5 file digest for FASTQ is for the file itself and not its content.
  • Fixed relative path handling in writers
  • FASTQ writing honors buffer parameters

Added

  • HDFS support for input and output. The URI should specify the schema (hdfs) and the host.
  • Support for convert any input to the Distmap format.
  • Common arguments

Changed

  • Removed legacy tools
  • Barcode metrics improvements: add header and PCT_RECORDS
  • Changed argument for split by read group, sample and/or library
  • Changed separator for barcode qualities if multiple indexes to space. Hyphen was used before, but it is a valid quality character.
  • (Developer) Removed legacy IO system
  • (Developer) Removed some deprecated code
  • (Developer) Repackaged some classes
  • (Developer) Removed GATKReadFilterPluginDescriptor hack
  • (Developer) Improve gradle build versioning

0.3.0 - 2017-02-06

Added

  • New tools for quality checking/standardizing: QualityEncodingDetector and StandardizeReads
  • New tool for converting to other format: ReadsToFastq
  • New tool for barcode detection: AssignReadGroupByBarcode
  • New tool for trimming/filtering pipeline: TrimReads
  • Input formats for every tool (except specified) includes BAM/SAM/CRAM/FASTQ
  • Tools output is in a consistent BAM/SAM/CRAM format (except for conversion tools)
  • Walker framework for single/pair-end reads traversal
  • BAM/SAM/CRAM/FASTQ sources are managed in the same way for a consistent output

Fixed

  • Fixed issue for CRAM files detection by extension and IO
  • Fixed barcode detection metrics (solves issue #77 and other minor issues)
  • BarcodeDictionary.getSampleReadGroups() returns an unmodifiable list

Changed

  • Tool deprecation: QualityChecker, StandardizeQuality, BamBarcodeDetector, FastqBarcodeDetector, TaggedBamToFastq, TrimFastq
  • Barcode file format (for new tools): strict tab-delimited file and header for each column (Picard's ExtractIlluminaBarcodes style). Requires at least a sample/barcode name and a barcode sequence. Library is optional.
  • Metrics from barcode detection ordered as the input file (samples and barcodes).
  • For new barcode tools, several indexes are joined with hyphen instead of underscore.
  • Default values are provided in the command line for list arguments in BarcodeArgumentCollection. If the user provide an option, the arguments will be overridden.
  • TrimReads applies trimmers before filters (see #101)

0.2.3 - 2016-11-01

Fixed

  • Validation stringency parameter (solves issue #35)
  • Option --forceOverwrite for force overwrite output (solves issue #39)
  • Option --readNameEncoding for Illumina/Casava read name formats (solves issue #48)

Changed

  • Testing is done with TestNG (developer)
  • Code cleaning and refactoring using GATK4 (developer improvement).

Added

  • Integration tests for the all the tools
  • Support for CRAM files reading (requires reference sequence). This support is not tested and relies on HTSJDK implementation
  • Option for create MD5 digest files from outputs (FATSQ/BAM/SAM/CRAM)

0.2.2 - 2016-08-09

Fixed

  • Option for allow higher qualities (-ahq) when standard encoding (issue #24)
  • Fixed bug in discarded output from TaggedBamToFastq for pair-end data (issue #34)

Changed

  • Change from Maven to Gradle as build system (developer improvement)

0.2.1 - 2016-04-19

Fixed

  • Fixed bug with standard qualities encoded withe the Illumina 1.8 protocol (issue #21)

Added

  • Maximum length option for TrimFastq (--max)
  • More informative error when quality checking fails

0.2.0 - 2016-03-21

Fixed

  • Fixed BAM file not standardize output when -nstd option is provided in BamBarcodeDetector
  • Fixed bug in (issue #19)

Changed

  • New barcode file format (without header): SampleName, Library, FirstBarcode, SecondBarcode. The program detects if the library is single or double indexed depending on the existence of the SecondBarcode column.
  • Now barcodes with several indexes are not merged in the read name, but separated by "_". This breaks the compatibility with respect to version 0.1.*

Added

  • Checking quality of every base when converting and every 1000 reads when not
  • Default values (BC and B2) for tags containing the barcodes in unmapped BAM files

0.1.4 - 2015-12-10

Fixed

  • Fixed bug when splitting with BamBarcodeDetector (issue #14)

0.1.3 - 2015-12-09

Changed

  • Logic of bacode detection by default: at least one difference between the best barcode and the second best is needed to do not discard a barcode. For the previous behaviour, use -d 0
  • TaggedBamToFastq check the PF flag in the BAM file and ignore reads that does not pass the vendor quality

Added

  • BamBarcodeDetector (new tool) for detect barcodes in a BAM file, adding read groups and split if requested
  • New options for barcode detection: maximum Ns, N not counting as mismatch, distance between best barcode and second barcode
  • Output metrics file when detecting barcodes
  • Barcode files are white space delimited (either space or tabs)

0.1.2 - 2015-11-26

Added

  • Checking output file existence
  • Create directories for output if they do not exists
  • Checking number of arguments

0.1.1 - 2015-11-24

Fixed

  • Fixed bug in single-end TrimFastq tool (issue #9)

0.1.0 - 2015-11-20

First pre-release