Merge pull request #53 from mskcc/release/0.2.0

Release/0.2.0
mskcc · Jul 24, 2019 · fa8351b · fa8351b
2 parents cab9ff4 + e2d72d2
commit fa8351b
Show file tree

Hide file tree

Showing 52 changed files with 4,176 additions and 0 deletions.
diff --git a/.gitignore b/.gitignore
@@ -0,0 +1,114 @@
+# Byte-compiled / optimized / DLL files
+__pycache__/
+*.py[cod]
+*$py.class
+
+# C extensions
+*.so
+
+# Distribution / packaging
+.Python
+build/
+develop-eggs/
+dist/
+downloads/
+eggs/
+.eggs/
+lib/
+lib64/
+parts/
+sdist/
+var/
+wheels/
+*.egg-info/
+.installed.cfg
+*.egg
+MANIFEST
+
+# PyInstaller
+#  Usually these files are written by a python script from a template
+#  before PyInstaller builds the exe, so as to inject date/other infos into it.
+*.manifest
+*.spec
+
+# Installer logs
+pip-log.txt
+pip-delete-this-directory.txt
+
+# Unit test / coverage reports
+htmlcov/
+.tox/
+.coverage
+.coverage.*
+.cache
+nosetests.xml
+coverage.xml
+*.cover
+.hypothesis/
+.pytest_cache/
+
+# Translations
+*.mo
+*.pot
+
+# Django stuff:
+*.log
+local_settings.py
+db.sqlite3
+
+# Flask stuff:
+instance/
+.webassets-cache
+
+# Scrapy stuff:
+.scrapy
+
+# Sphinx documentation
+docs/_build/
+
+# PyBuilder
+target/
+
+# Jupyter Notebook
+.ipynb_checkpoints
+
+# pyenv
+.python-version
+
+# celery beat schedule file
+celerybeat-schedule
+
+# SageMath parsed files
+*.sage.py
+
+# Environments
+.env
+.venv
+env/
+venv/
+ENV/
+env.bak/
+venv.bak/
+
+# Spyder project settings
+.spyderproject
+.spyproject
+
+# Rope project settings
+.ropeproject
+
+# mkdocs documentation
+/site
+
+# mypy
+.mypy_cache/
+
+# Mac
+.DS_Store
+
+# vscode
+.vscode/*
+!.vscode/settings.json
+!.vscode/tasks.json
+!.vscode/launch.json
+!.vscode/extensions.json
diff --git a/abra2_2.17/README.md b/abra2_2.17/README.md
@@ -0,0 +1,21 @@
+# CWL and Dockerfile for running ABRA2
+
+## Version of tools in docker image (/container/Dockerfile)
+
+| Tool	| Version	| Location	|
+|---	|---	|---	|
+| openjdk  	| 8  	|  -	|
+| ABRA2  	| 2.17	|  https://github.com/mozack/abra2/releases/download/v2.19/abra2-2.17.jar	|
+
+[![](https://images.microbadger.com/badges/version/mskcc/abra2:0.1.0.svg)](https://microbadger.com/images/mskcc/abra2:0.1.0 "Get your own version badge on microbadger.com") [![](https://images.microbadger.com/badges/image/mskcc/abra2:0.1.0.svg)](https://microbadger.com/images/mskcc/abra2:0.1.0 "Get your own image badge on microbadger.com") [![](https://images.microbadger.com/badges/license/mskcc/abra2:0.1.0.svg)](https://microbadger.com/images/mskcc/abra2:0.1.0 "Get your own license badge on microbadger.com")
+
+## CWL
+
+- CWL specification 1.0
+- Use example_inputs.yaml to see the inputs to the cwl
+- Example Command using [toil](https://toil.readthedocs.io):
+
+```bash
+    > toil-cwl-runner abra2_2.17.cwl example_inputs.yaml
+```
+
diff --git a/abra2_2.17/abra2_2.17.cwl b/abra2_2.17/abra2_2.17.cwl
@@ -0,0 +1,198 @@
+class: CommandLineTool
+cwlVersion: v1.0
+$namespaces:
+  dct: 'http://purl.org/dc/terms/'
+  doap: 'http://usefulinc.com/ns/doap#'
+  foaf: 'http://xmlns.com/foaf/0.1/'
+id: abra2_2.17
+baseCommand:
+  - java
+inputs:
+  - id: memory_per_job
+    type: int?
+    doc: Memory per job in megabytes
+  - id: memory_overhead
+    type: int?
+    doc: Memory overhead per job in megabytes
+  - id: number_of_threads
+    type: int?
+    inputBinding:
+      position: 0
+      prefix: '--threads'
+  - id: input_bam
+    type:
+      - File
+      - type: array
+        items: File
+    inputBinding:
+      position: 0
+      prefix: '--in'
+    doc: Required list of input sam or bam file (s) separated by comma
+    secondaryFiles:
+      - ^.bai
+  - id: working_directory
+    type: Directory?
+    inputBinding:
+      position: 0
+      prefix: '--tmpdir'
+    doc: Set the temp directory (overrides java.io.tmpdir)
+  - id: reference_fasta
+    type: File
+    inputBinding:
+      position: 0
+      prefix: '--ref'
+    doc: Genome reference location
+    secondaryFiles:
+      - ^.fai
+  - id: targets
+    type: File
+    inputBinding:
+      position: 0
+      prefix: '--targets'
+  - id: kmer_size
+    type: string?
+    inputBinding:
+      position: 0
+      prefix: '--kmer'
+    doc: >-
+      Optional assembly kmer size(delimit with commas if multiple sizes
+      specified)
+  - id: maximum_average_depth
+    type: int?
+    inputBinding:
+      position: 0
+      prefix: '--mad'
+    doc: >-
+      Regions with average depth exceeding this value will be downsampled
+      (default: 1000)
+  - id: soft_clip_contig
+    type: string?
+    inputBinding:
+      position: 0
+      prefix: '--sc'
+    doc: >-
+      Soft clip contig args
+      [max_contigs,min_base_qual,frac_high_qual_bases,min_soft_clip_len]
+      (default:16,13,80,15)
+  - id: maximum_mixmatch_rate
+    type: float?
+    inputBinding:
+      position: 0
+      prefix: '--mmr'
+    doc: >-
+      Max allowed mismatch rate when mapping reads back to contigs (default:
+      0.05)
+  - id: scoring_gap_alignments
+    type: string?
+    inputBinding:
+      position: 0
+      prefix: '--sga'
+    doc: >-
+      Scoring used for contig alignments(match,
+      mismatch_penalty,gap_open_penalty,gap_extend_penalty) (default:8,32,48,1)
+  - id: contig_anchor
+    type: string?
+    inputBinding:
+      position: 0
+      prefix: '--ca'
+    doc: >-
+      Contig anchor [M_bases_at_contig_edge,max_mismatches_near_edge]
+      (default:10,2)
+  - id: window_size
+    type: string?
+    inputBinding:
+      position: 0
+      prefix: '--ws'
+    doc: |-
+      Processing window size and overlap
+      (size,overlap) (default: 400,200)
+  - id: consensus_sequence
+    type: boolean?
+    inputBinding:
+      position: 0
+      prefix: '--cons'
+    doc: Use positional consensus sequence when aligning high quality soft clipping
+  - id: output_bams
+    type:
+      - string
+      - type: array
+        items: string
+    inputBinding:
+      position: 0
+      prefix: '--out'
+    doc: Required list of output sam or bam file (s) separated by comma
+  - id: ignore_bad_assembly
+    type: boolean?
+    inputBinding:
+      position: 0
+      prefix: '--ignore-bad-assembly'
+    doc: Use this option to avoid parsing errors for corrupted assemblies
+  - id: bam_index
+    type: boolean?
+    inputBinding:
+      position: 0
+      prefix: '--index'
+    doc: >-
+      Enable BAM index generation when outputting sorted alignments (may require
+      additonal memory)
+  - id: input_vcf
+    type: File?
+    inputBinding:
+      position: 0
+      prefix: '--in-vcf'
+    doc: >-
+      VCF containing known (or suspected) variant sites.  Very large files
+      should be avoided.
+  - id: no_sort
+    type: boolean?
+    inputBinding:
+      position: 0
+      prefix: '--nosort'
+    doc: Do not attempt to sort final output
+outputs:
+  - id: realigned_bam
+    type:
+      - 'null'
+      - File
+      - type: array
+        items: File
+    outputBinding:
+      glob: |
+        *abra.bam
+    secondaryFiles:
+      - ^.bai
+label: abra2_2.17
+arguments:
+  - position: 0
+    valueFrom: "${\n  if(inputs.memory_per_job && inputs.memory_overhead) {\n   \n    if(inputs.memory_per_job % 1000 == 0) {\n    \t\n      return \"-Xmx\" + (inputs.memory_per_job/1000).toString() + \"G\"\n    }\n    else {\n      \n      return \"-Xmx\" + Math.floor((inputs.memory_per_job/1000)).toString() + \"G\" \n    }\n  }\n  else if (inputs.memory_per_job && !inputs.memory_overhead){\n    \n    if(inputs.memory_per_job % 1000 == 0) {\n    \t\n      return \"-Xmx\" + (inputs.memory_per_job/1000).toString() + \"G\"\n    }\n    else {\n      \n      return \"-Xmx\" + Math.floor((inputs.memory_per_job/1000)).toString() + \"G\" \n    }\n  }\n  else if(!inputs.memory_per_job && inputs.memory_overhead){\n    \n    return \"-Xmx15G\"\n  }\n  else {\n    \n  \treturn \"-Xmx15G\"\n  }\n}"
+  - position: 0
+    prefix: '-jar'
+    valueFrom: /usr/local/bin/abra2.jar
+requirements:
+  - class: ResourceRequirement
+    ramMin: "${\r  if(inputs.memory_per_job && inputs.memory_overhead) {\r   \r    return inputs.memory_per_job + inputs.memory_overhead\r  }\r  else if (inputs.memory_per_job && !inputs.memory_overhead){\r    \r   \treturn inputs.memory_per_job + 2000\r  }\r  else if(!inputs.memory_per_job && inputs.memory_overhead){\r    \r    return 15000 + inputs.memory_overhead\r  }\r  else {\r    \r  \treturn 17000 \r  }\r}"
+    coresMin: "${\r  if (inputs.number_of_threads) {\r    \r   \treturn inputs.number_of_threads \r  }\r  else {\r    \r    return 4\r  }\r}"
+  - class: DockerRequirement
+    dockerPull: 'mskcc/abra2:0.1.0'
+  - class: InlineJavascriptRequirement
+'dct:contributor':
+  - class: 'foaf:Organization'
+    'foaf:member':
+      - class: 'foaf:Person'
+        'foaf:mbox': 'mailto:shahr2@mskcc.org'
+        'foaf:name': Ronak Shah
+    'foaf:name': Memorial Sloan Kettering Cancer Center
+'dct:creator':
+  - class: 'foaf:Organization'
+    'foaf:member':
+      - class: 'foaf:Person'
+        'foaf:mbox': 'mailto:shahr2@mskcc.org'
+        'foaf:name': Ronak Shah
+    'foaf:name': Memorial Sloan Kettering Cancer Center
+'doap:release':
+  - class: 'doap:Version'
+    'doap:name': abra2
+    'doap:revision': 2.17
+  - class: 'doap:Version'
+    'doap:name': cwl-wrapper
+    'doap:revision': 1.0.0
diff --git a/abra2_2.17/container/Dockerfile b/abra2_2.17/container/Dockerfile
@@ -0,0 +1,40 @@
+################## BASE IMAGE ######################
+
+FROM openjdk:8
+
+################## ARGUMENTS########################
+
+ARG BUILD_DATE
+ARG BUILD_VERSION
+ARG LICENSE="Apache-2.0"
+ARG JAVA_VERSION=8
+ARG ABRA2_VERSION=2.17
+
+################## METADATA ########################
+
+LABEL org.opencontainers.image.vendor="MSKCC"
+LABEL org.opencontainers.image.authors="Ronak H Shah ([email protected])" 
+
+LABEL org.opencontainers.image.created=${BUILD_DATE} \ 
+	org.opencontainers.image.version=${BUILD_VERSION} \
+	org.opencontainers.image.licenses=${LICENSE} \
+	org.opencontainers.image.version.java=${JAVA_VERSION} \
+	org.opencontainers.image.version.abra2=${ABRA2_VERSION} \
+	org.opencontainers.image.source.abra2="https://github.com/mozack/abra2/releases/"
+
+LABEL org.opencontainers.image.description="This container uses OPENJDK ${JAVA_VERSION} as the base image to build abra2 version ${ABRA2_VERSION}"
+
+################## INSTALL ##########################
+
+WORKDIR /usr/src
+
+RUN apt-get update && \
+	apt-get --no-install-recommends install -y \
+	wget && \
+	apt-get clean autoclean && \
+	apt-get autoremove -y && \
+	rm -rf /var/lib/apt/lists/* /tmp/* /var/tmp/*
+
+RUN wget "https://github.com/mozack/abra2/releases/download/v${ABRA2_VERSION}/abra2-${ABRA2_VERSION}.jar" && \
+	chmod 755 /usr/src/abra2-${ABRA2_VERSION}.jar && \
+    cp -s /usr/src/abra2-${ABRA2_VERSION}.jar /usr/local/bin/abra2.jar