==================
Changed
-------
- **BACKWARD INCOMPATIBLE:** Remove Ubuntu 17.04 base Docker image since it has
has reached its end of life and change all images to use the new ubuntu 17.10
base image
- **BACKWARD INCOMPATIBLE:** Require ``species`` and ``build`` inputs in the
following processes:
- ``upload-genome``
- ``upload-gtf``
- ``upload-gff3``
- ``upload-bam``
- ``upload-bam-indexed``
- **BACKWARD INCOMPATIBLE:** Track ``species`` and ``build`` information in the
following processes:
- ``cuffmerge``
- alignment processes
- variant calling processes
- JBrowse processes
- **BACKWARD INCOMPATIBLE:** Track ``species``, ``build`` and ``feature_type``
in the following processes:
- ``upload-expression-star``
- quantification processes
- differential expression processes
- **BACKWARD INCOMPATIBLE:** Track ``species`` in gene set (Venn) and
``goenrichment`` processes
- **BACKWARD INCOMPATIBLE:** Rename ``genes_source`` input to ``source`` in
hierarchical clustering and PCA processes
- **BACKWARD INCOMPATIBLE:** Remove the following obsolete processes:
- Dictyostelium-specific ncRNA quantification
- ``go-geneset``
- bayseq differential expression
- ``cuffmerge-gtf-to-gff3``
- ``transdecoder``
- ``web-gtf-dictybase``
- ``upload-rmsk``
- ``snpdat``
- **BACKWARD INCOMPATIBLE:** Unify output fields of processes of type
``data:annotation``
- **BACKWARD INCOMPATIBLE:** Rename the organism field names to species in
``rna-seq`` and ``cutadapt-star-htseq`` descriptor schemas
- **BACKWARD INCOMPATIBLE:** Rename the ``genome_and_annotation`` field name
to ``species`` in ``bcm-*`` descriptor schemas and use the full species name
for the ``species`` field values
- **BACKWARD INCOMPATIBLE:** Refactor ``featureCounts`` process
- **BACKWARD INCOMPATIBLE:** Change ``import-sra`` process to work with
``resolwebio/utils`` Docker image and refactor its inputs
- Require Resolwe 7.x
- Add environment export for Jenkins so that the manager will use a
globally-unique channel name
- Set ``scheduling_class`` of gene and sample hierarchical clustering processes
to ``interactive``
- Change base Docker images of ``resolwebio/rnaseq`` and ``resolwebio/dnaseq``
to ``resolwebio/base:ubuntu-18.04``
- Use the latest versions of the following Python packages in
``resolwebio/rnaseq`` Docker image: Cutadapt 1.15, Apache Arrow 0.8.0,
pysam 0.13, and xopen 0.3.2
- Use the latest versions of the following Python packages in
``resolwebio/dnaseq`` Docker image: Bokeh 0.12.13, pandas 0.22.0,
Matplotlib 2.1.2, six 1.11.0, PyYAML 3.12, Jinja2 2.10, NumPy 1.14.0,
Tornado 4.5.3, and pytz 2017.3
- Use the latest version of ``wigToBigWig`` tool in ``resolwebio/chipseq``
Docker image
- Use ``resolwebio/rnaseq:3.0.0`` Docker image in ``goenrichment``,
``upload-gaf`` and ``upload-obo`` processes
- Use ``resolwebio/dnaseq:3.0.0`` Docker image in ``filtering_chemut`` process
- Change ``cuffnorm`` process type to ``data:cuffnorm``
- Set type of ``coverage-garvan`` process to ``data:exomecoverage``
- Remove ``gsize`` input from ``macs14`` process and automate genome size
selection
- Adjust ``bam-split`` process so it can be included in workflows
- Make ID attribute labels in ``featureCounts`` more informative
- Change 'source' to 'gene ID database' in labes and descriptions
- Change ``archive-samples`` process to create different IGV session files for
``build`` and ``species``
- Expose advanced parameters in Chemical Mutagenesis workflow
- Clarify some descriptions in the ``filtering_chemut`` process and ``chemut``
workflow
- Change expected genome build formatting for hybrid genomes in ``bam-split``
process
- Set the ``cooksCutoff`` parameter to ``FALSE`` in ``deseq.R`` tool
- Rename 'Expressions (BCM)' to 'Dicty expressions'
Added
-----
- Mechanism to override the manager's control channel prefix from the
environment
- Add Ubuntu 17.10 and Ubuntu 18.04 base Docker images
- Add ``resolwebio/utils`` Docker image
- Add ``BBMap``, ``Trimmomatic``, ``Subread``, ``Salmon``, and
``dexseq_prepare_annotation2`` tools and ``DEXSeq`` and ``loadSubread`` R
libraries to ``resolwebio/rnaseq`` Docker image
- Add abstract processes that ensure that all processes that inherit from them
have the input and output fields that are defined in them:
- ``abstract-alignment``
- ``abstract-annotation``
- ``abstract-expression``
- ``abstract-differentialexpression``
- ``abstract-bed``
- Add miRNA workflow
- Add ``prepare-geo-chipseq`` and ``prepare-geo-rnaseq`` processes that produce
a tarball with necessary data and folder structure for GEO upload
- Add ``library-strandedness`` process which uses the ``Salmon`` tool built-in
functionality to detect the library strandedness information
- Add ``species`` and ``genome build`` output fields to ``macs14`` process
- Expose additional parameters in ``alignment-star``, ``cutadapt-single`` and
``cutadapt-paired`` processes
- Add ``merge expressions`` to ``archive-samples`` process
- Add description of batch mode to Expression aggregator process
- Add error and warning messages to the ``cuffnorm`` process
- Add optional ``species`` input to hierarchical clustering and PCA processes
- Add Rattus norvegicus species choice to the ``rna-seq`` descriptor schema
to allow running RNA-seq workflow for this species from the Recipes
Fixed
-----
- Fix custom argument passing script for ``Trimmomatic`` in
``resolwebio/rnaseq`` Docker image
- Fix installation errors for ``dexseq-prepare-annotation2`` in
``resolwebio/rnaseq`` Docker image
- Fix ``consensus_subreads`` input option in Subread process
- Limit variant-calling process in the chemical mutagenesis workflow and the
Picard tools run inside to 16 GB of memory to prevent them from crashing
because they try to use too much memory
- The chemical mutagenesis workflow was erroneously categorized as
``data:workflow:rnaseq:cuffquant`` type. This is switched to
``data:workflow:chemut`` type.
- Fix handling of NA values in Differential expression results table. NA values
were incorrectly replaced with value 0 instead of 1
- Fix ``cuffnorm`` process to work with samples containing dashes in
their name and dispense prefixing sample names starting with numbers
with 'X' in the ``cuffnorm`` normalization outputs
- Fix ``cuffnorm`` process' outputs to correctly track species and
build information
- Fix typos and sync parameter description common to ``featureCounts``
and ``miRNA`` workflow
==================