@prefix LoadListingRequirement: .
@prefix CommandLineBinding: .
@prefix xsd: .
@prefix DockerRequirement: .
@prefix Workflow: .
@prefix sld: .
@prefix rdfs: .
@prefix CommandOutputBinding: .
@prefix ns1: .
@prefix NetworkAccess: .
@prefix rdf: .
@prefix ScatterMethod: .
@prefix cwl: .
@prefix CommandLineTool: .
a cwl:CommandLineTool ;
rdfs:comment "Extracts best fragment length from SPP output text file" ;
ns1:original_cwlVersion "v1.0" ;
cwl:baseCommand ( "extract-best-frag-length.sh" ) ;
cwl:cwlVersion ;
cwl:hints [ a ;
NetworkAccess:networkAccess true
] ;
cwl:hints [ a cwl:DockerRequirement ;
DockerRequirement:dockerPull "reddylab/workflow-utils:ggr"
] ;
cwl:hints [ a ;
LoadListingRequirement:loadListing
"deep_listing"
] ;
cwl:inputs ;
cwl:outputs ;
cwl:stdout "best_frag_length" .
rdfs:comment "Effective genome size. It can be 1.0e+9 or 1000000000, or shortcuts:'hs' for human (2.7e9), 'mm' for mouse (1.87e9), 'ce' for C. elegans (9e7) and 'dm' for fruitfly (1.2e8), Default:hs." ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "-g"
] ;
sld:type xsd:string , sld:null .
rdfs:comment "Two column tab-delimited file with chromosome size information" ;
sld:type cwl:File .
rdfs:comment "Peak counts within replicate" ;
cwl:outputSource ;
sld:type [ sld:items cwl:File ;
sld:type sld:array
] .
cwl:in , , , , , , , ;
cwl:out , , , ;
cwl:run ;
cwl:scatter ( ) ;
cwl:scatterMethod ScatterMethod:dotproduct .
rdfs:comment "-filtchr= , Pattern to use to remove tags that map to specific chromosomes e.g. _ will remove all tags that map to chromosomes with _ in their name" ;
cwl:inputBinding [ CommandLineBinding:prefix "-filtchr=" ;
CommandLineBinding:separate false
] ;
sld:type xsd:string , sld:null .
a cwl:CommandLineTool ;
ns1:original_cwlVersion "v1.0" ;
cwl:arguments ( [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "-n" ;
cwl:valueFrom "$(inputs.treatment[0].path.replace(/^.*[\\\\\\/]/, '').replace(/\\.[^/.]+$/, ''))"
]
) ;
cwl:baseCommand ( "macs2" "callpeak" ) ;
cwl:cwlVersion ;
cwl:hints [ a ;
LoadListingRequirement:loadListing
"deep_listing"
] ;
cwl:hints [ a ;
NetworkAccess:networkAccess true
] ;
cwl:hints [ a cwl:DockerRequirement ;
DockerRequirement:dockerPull "dukegcb/macs2"
] ;
cwl:inputs , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ;
cwl:outputs , , , ;
cwl:requirements [ a cwl:InlineJavascriptRequirement ] .
rdfs:comment "peakshift/phantomPeak extended fragment results file" ;
cwl:outputSource ;
sld:type [ sld:items cwl:File ;
sld:type sld:array
] .
a cwl:CommandLineTool ;
rdfs:comment "Filter BAM file to only include reads overlapping with a BED file" ;
ns1:original_cwlVersion "v1.0" ;
cwl:baseCommand ( "samtools" "view" "-b" "-h" ) ;
cwl:cwlVersion ;
cwl:hints [ a cwl:DockerRequirement ;
DockerRequirement:dockerPull "dukegcb/samtools:1.3"
] ;
cwl:hints [ a ;
NetworkAccess:networkAccess true
] ;
cwl:hints [ a ;
LoadListingRequirement:loadListing
"deep_listing"
] ;
cwl:inputs , ;
cwl:outputs ;
cwl:requirements [ a cwl:InlineJavascriptRequirement ] ;
cwl:stdout "$(inputs.input_bam_file.path.replace(/^.*[\\\\\\/]/, '') + '.in_peaks.bam')" .
a cwl:CommandLineTool ;
rdfs:comment "Count number of dedup-ed reads used in peak calling" ;
ns1:original_cwlVersion "v1.0" ;
cwl:baseCommand ( "count-filtered-reads-macs2.sh" ) ;
cwl:cwlVersion ;
cwl:hints [ a ;
LoadListingRequirement:loadListing
"deep_listing"
] ;
cwl:hints [ a cwl:DockerRequirement ;
DockerRequirement:dockerPull "reddylab/workflow-utils:ggr"
] ;
cwl:hints [ a ;
NetworkAccess:networkAccess true
] ;
cwl:inputs ;
cwl:outputs ;
cwl:requirements [ a cwl:InlineJavascriptRequirement ] ;
cwl:stdout "$(inputs.peak_xls_file.path.replace(/^.*[\\\\\\/]/, '').replace(/\\_peaks\\.xls$/, '_read_count.txt'))" .
cwl:valueFrom ".peak_count.within_replicate.txt" .
cwl:valueFrom "bed6+4" .
rdfs:comment "Peaks in bigBed format" ;
cwl:outputSource ;
sld:type [ sld:items cwl:File ;
sld:type sld:array
] .
rdfs:comment "Control sample file." ;
cwl:inputBinding [ CommandLineBinding:position 2 ;
CommandLineBinding:prefix "--control"
] ;
sld:type cwl:File , sld:null .
rdfs:comment "-savd= OR -savd, save Rdata file" ;
cwl:inputBinding [ cwl:valueFrom "${ if (self) return \"-savd=\" + inputs.input_bam.path.replace(/^.*[\\\\\\/]/, \"\").replace(/\\.[^/.]+$/, \"\") + \".spp.Rdata\"; return null}" ] ;
sld:type xsd:boolean , sld:null .
rdfs:comment "Rdata file from the run_spp.R run" ;
cwl:outputBinding [ CommandOutputBinding:glob
"$(inputs.input_bam.path.replace(/^.*[\\\\\\/]/, \"\").replace(/\\.[^/.]+$/, \"\") + \".spp.Rdata\")" ] ;
sld:type cwl:File , sld:null .
cwl:default ".bb" ;
sld:type xsd:string .
cwl:source .
rdfs:comment ", full path and name (or URL) of tagAlign/BAM file (can be gzipped) (FILE EXTENSION MUST BE tagAlign.gz, tagAlign, bam or bam.gz)" ;
cwl:inputBinding [ CommandLineBinding:prefix "-i=" ;
CommandLineBinding:separate false
] ;
sld:type cwl:File , sld:null .
a cwl:CommandLineTool ;
ns1:original_cwlVersion "v1.0" ;
cwl:arguments ( [ cwl:valueFrom "$(\"-out=\" + inputs.input_bam.path.replace(/^.*[\\\\\\/]/, \"\").replace(/\\.[^/.]+$/, \"\") + \".spp_cross_corr.txt\")" ]
[ CommandLineBinding:shellQuote false ;
cwl:valueFrom "$(\"-tmpdir=\"+runtime.tmpdir)"
]
) ;
cwl:baseCommand ( "run_spp.R" ) ;
cwl:cwlVersion ;
cwl:hints [ a cwl:DockerRequirement ;
DockerRequirement:dockerPull "dukegcb/spp"
] ;
cwl:hints [ a ;
LoadListingRequirement:loadListing
"deep_listing"
] ;
cwl:hints [ a ;
NetworkAccess:networkAccess true
] ;
cwl:inputs , , , , , , , , , , , , , ;
cwl:outputs , , , , ;
cwl:requirements [ a cwl:InlineJavascriptRequirement ] ;
cwl:requirements [ a cwl:ShellCommandRequirement ] .
rdfs:comment "If set, MACS will try to call broad peaks by linking nearby highly enriched regions. The linking region is controlled by another cutoff through --linking-cutoff. The maximum linking region length is 4 times of d from MACS. DEFAULT: False " ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "--broad"
] ;
sld:type xsd:boolean , sld:null .
rdfs:comment "-type=bedN[+[P]] :\n N is between 3 and 15,\n optional (+) if extra \"bedPlus\" fields,\n optional P specifies the number of extra fields. Not required, but preferred.\n Examples: -type=bed6 or -type=bed6+ or -type=bed6+3\n (see http://genome.ucsc.edu/FAQ/FAQformat.html#format1)\n" ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "-type=" ;
CommandLineBinding:separate false
] ;
sld:type xsd:string , sld:null .
rdfs:comment "While set, MACS2 will analyze number or total length of peaks that can be called by different p-value cutoff then output a summary table to help user decide a better cutoff. The table will be saved in NAME_cutoff_analysis.txt file. Note, minlen and maxgap may affect the results. WARNING: May take ~30 folds longer time to finish. DEFAULT: False Post-processing options: " ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "--cutoff-analysis"
] ;
sld:type xsd:boolean , sld:null .
cwl:outputBinding [ CommandOutputBinding:glob
"$(inputs.peak_xls_file.path.replace(/^.*[\\\\\\/]/, '').replace(/\\_peaks\\.xls$/, '_read_count.txt'))" ] ;
sld:type cwl:File .
cwl:valueFrom "BAM" .
rdfs:comment "Filtered aligned BAM file by BED coordinates file" ;
cwl:outputBinding [ CommandOutputBinding:glob
"$(inputs.input_bam_file.path.replace(/^.*[\\\\\\/]/, '') + '.in_peaks.bam')" ] ;
sld:type cwl:File .
cwl:source .
cwl:valueFrom "${return true}" .
rdfs:comment "MFOLD MFOLD, --mfold MFOLD MFOLD\nSelect the regions within MFOLD range of high-\nconfidence enrichment ratio against background to\nbuild model. Fold-enrichment in regions must be lower\nthan upper limit, and higher than the lower limit. Use\nas \"-m 10 30\". DEFAULT:5 50\n" ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "-m"
] ;
sld:type xsd:string , sld:null .
rdfs:comment "Treatment sample file(s). If multiple files are given as -t A B C, then they will all be read and pooled together. IMPORTANT: the first sample will be used as the outputs basename." ;
cwl:inputBinding [ CommandLineBinding:position 2 ;
CommandLineBinding:prefix "--treatment"
] ;
sld:type [ sld:items cwl:File ;
sld:type sld:array
] .
cwl:valueFrom "${return 200}" .
cwl:source .
rdfs:comment "BW Band width for picking regions to compute fragment\nsize. This value is only used while building the\nshifting model. DEFAULT: 300\n" ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "--bw"
] ;
sld:type xsd:int , sld:null .
rdfs:comment "KEEPDUPLICATES\nIt controls the MACS behavior towards duplicate tags\nat the exact same location -- the same coordination\nand the same strand. The 'auto' option makes MACS\ncalculate the maximum tags at the exact same location\nbased on binomal distribution using 1e-5 as pvalue\ncutoff; and the 'all' option keeps every tags. If an\ninteger is given, at most this number of tags will be\nkept at the same location. The default is to keep one\ntag at the same location. Default: 1\n" ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "--keep-dup"
] ;
sld:type xsd:string , sld:null .
cwl:outputBinding [ CommandOutputBinding:glob
"$(inputs.peaks.path.replace(/^.*[\\\\\\/]/, '').replace(/\\.([^/.]+)$/, \"\\.trunked_scores\\.$1\"))" ] ;
sld:type cwl:File .
a cwl:CommandLineTool ;
rdfs:comment "Extract mapped reads from BAM file using Samtools flagstat command" ;
ns1:original_cwlVersion "v1.0" ;
cwl:arguments ( [ CommandLineBinding:position 10000 ;
CommandLineBinding:shellQuote false ;
cwl:valueFrom " | head -n1 | cut -f 1 -d ' '"
]
) ;
cwl:baseCommand ( "samtools" "flagstat" ) ;
cwl:cwlVersion ;
cwl:hints [ a cwl:DockerRequirement ;
DockerRequirement:dockerPull "dukegcb/samtools"
] ;
cwl:hints [ a ;
LoadListingRequirement:loadListing
"deep_listing"
] ;
cwl:hints [ a ;
NetworkAccess:networkAccess true
] ;
cwl:inputs , ;
cwl:outputs ;
cwl:requirements [ a cwl:ShellCommandRequirement ] ;
cwl:requirements [ a cwl:InlineJavascriptRequirement ] ;
cwl:stdout "$(inputs.input_bam_file.path.replace(/^.*[\\\\\\/]/, '').replace(/\\.[^/.]+$/, \"\") + inputs.output_suffix)" .
rdfs:comment "If True, MACS will use fixed background lambda as local lambda for every peak region. Normally, MACS calculates a dynamic local lambda to reflect the local bias due to potential chromatin structure. " ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "--nolambda"
] ;
sld:type xsd:boolean , sld:null .
rdfs:comment "-as=fields.as - If you have non-standard \"bedPlus\" fields, it's great to put a definition\n of each field in a row in AutoSql format here.1)\n" ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "-as=" ;
CommandLineBinding:separate false
] ;
sld:type cwl:File , sld:null .
rdfs:comment "peakshift/phantomPeak results file" ;
cwl:outputSource ;
sld:type [ sld:items cwl:File ;
sld:type sld:array
] .
rdfs:comment "SEED Set the random seed while down sampling data. Must be\na non-negative integer in order to be effective.\nDEFAULT: not set\n" ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "--seed"
] ;
sld:type xsd:int , sld:null .
rdfs:comment "Samtools Flagstat report file" ;
cwl:outputBinding [ CommandOutputBinding:glob
"$(inputs.input_bam_file.path.replace(/^.*[\\\\\\/]/, '').replace(/\\.[^/.]+$/, \"\") + inputs.output_suffix)" ] ;
sld:type cwl:File .
cwl:in ;
cwl:out ;
cwl:run ;
cwl:scatter ( ) .
rdfs:comment "-speak=, user-defined cross-correlation peak strandshift" ;
cwl:inputBinding [ CommandLineBinding:prefix "-speak=" ;
CommandLineBinding:separate false
] ;
sld:type xsd:string , sld:null .
cwl:in , ;
cwl:out ;
cwl:run ;
cwl:scatter ( ) .
a cwl:Workflow ;
rdfs:comment "ATAC-seq 04 quantification - SE" ;
ns1:original_cwlVersion "v1.0" ;
Workflow:steps , , , , , , , , ;
cwl:cwlVersion ;
cwl:hints [ a ;
LoadListingRequirement:loadListing
"deep_listing"
] ;
cwl:hints [ a ;
NetworkAccess:networkAccess true
] ;
cwl:inputs , , , , ;
cwl:outputs , , , , , , , , , ;
cwl:requirements [ a cwl:ScatterFeatureRequirement ] ;
cwl:requirements [ a cwl:InlineJavascriptRequirement ] ;
cwl:requirements [ a cwl:StepInputExpressionRequirement ] .
cwl:in ;
cwl:out ;
cwl:run ;
cwl:scatter ( ) .
cwl:source ;
cwl:valueFrom "$([self])" .
cwl:in , , , ;
cwl:out ;
cwl:run ;
cwl:scatter ( ) .
cwl:in ;
cwl:out ;
cwl:run ;
cwl:scatter ( ) .
cwl:outputBinding [ CommandOutputBinding:glob "best_frag_length" ;
CommandOutputBinding:outputEval
"$(Number(self[0].contents.replace('\\n', '')))" ;
cwl:loadContents true
] ;
sld:type xsd:float .
rdfs:comment "Aligned BAM file to filter" ;
cwl:inputBinding [ CommandLineBinding:position
1 ] ;
sld:type cwl:File .
rdfs:comment "Pvalue cutoff for peak detection. DEFAULT: not set. -q, and -p are mutually exclusive. If pvalue cutoff is set, qvalue will not be calculated and reported as -1 in the final .xls file.." ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "-p"
] ;
sld:type xsd:float , sld:null .
rdfs:comment "Input bed file" ;
cwl:inputBinding [ CommandLineBinding:position
2 ] ;
sld:type cwl:File .
rdfs:comment "peakshift/phantomPeak results file" ;
cwl:outputSource ;
sld:type [ sld:items cwl:File ;
sld:type sld:array
] .
cwl:inputBinding [ CommandLineBinding:position
1 ] ;
sld:type cwl:File .
rdfs:comment "-savr= OR -savr RegionPeak file name (variable width peaks with regions of enrichment)" ;
cwl:inputBinding [ cwl:valueFrom "${ if (self) return \"-savr=\" + inputs.input_bam.path.replace(/^.*[\\\\\\/]/, \"\").replace(/\\.[^/.]+$/, \"\") + \".spp.regionPeak\"; return null}" ] ;
sld:type xsd:boolean , sld:null .
rdfs:comment "-f {AUTO,BAM,SAM,BED,ELAND,ELANDMULTI,ELANDEXPORT,BOWTIE,BAMPE}, --format {AUTO,BAM,SAM,BED,ELAND,ELANDMULTI,ELANDEXPORT,BOWTIE,BAMPE} Format of tag file, \"AUTO\", \"BED\" or \"ELAND\" or \"ELANDMULTI\" or \"ELANDEXPORT\" or \"SAM\" or \"BAM\" or \"BOWTIE\" or \"BAMPE\". The default AUTO option will let MACS decide which format the file is. Note that MACS can't detect \"BAMPE\" or \"BEDPE\" format with \"AUTO\", and you have to implicitly specify the format for \"BAMPE\" and \"BEDPE\". DEFAULT: \"AUTO\"." ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "-f"
] ;
sld:type xsd:string , sld:null .
rdfs:comment "Peaks summits bedfile." ;
cwl:outputBinding [ CommandOutputBinding:glob
"$(inputs.treatment[0].path.replace(/^.*[\\\\\\/]/, '').replace(/\\.[^/.]+$/, '') + '_summits.bed')" ] ;
sld:type cwl:File , sld:null .
rdfs:comment "The arbitrary extension size in bp. When nomodel is true, MACS will use this value as fragment size to extend each read towards 3' end, then pile them up. It's exactly twice the number of obsolete SHIFTSIZE. In previous language, each read is moved 5'->3' direction to middle of fragment by 1/2 d, then extended to both direction with 1/2 d. This is equivalent to say each read is extended towards 5'->3' into a d size fragment. DEFAULT: 200. EXTSIZE and SHIFT can be combined when necessary. Check SHIFT option." ;
cwl:inputBinding [ CommandLineBinding:position 1 ;
CommandLineBinding:prefix "--extsize"
] ;
sld:type xsd:float , sld:null .
rdfs:comment "Peaks information/report file." ;
cwl:outputBinding [ CommandOutputBinding:glob
"$(inputs.treatment[0].path.replace(/^.*[\\\\\\/]/, '').replace(/\\.[^/.]+$/, '') + '_peaks.xls')" ] ;
sld:type cwl:File .
a cwl:CommandLineTool ;
rdfs:comment "Counts lines in a file and returns a suffixed file with that number" ;
ns1:original_cwlVersion "v1.0" ;
CommandLineTool:stdin "$(inputs.input_file.path)" ;
cwl:baseCommand ( "wc" "-l" ) ;
cwl:cwlVersion ;
cwl:hints [ a ;
LoadListingRequirement:loadListing
"deep_listing"
] ;
cwl:hints [ a ;
NetworkAccess:networkAccess true
] ;
cwl:inputs , ;
cwl:outputs ;
cwl:requirements [ a cwl:InlineJavascriptRequirement ] ;
cwl:stdout "$(inputs.input_file.path.replace(/^.*[\\\\\\/]/, '') + inputs.output_suffix)" .
rdfs:comment "Effective genome size used by MACS2. It can be numeric or a shortcuts:'hs' for human (2.7e9), 'mm' for mouse (1.87e9), 'ce' for C. elegans (9e7) and 'dm' for fruitfly (1.2e8), Default:hs" ;
cwl:default "hs" ;
sld:type xsd:string .
rdfs:comment "peakshift/phantomPeak results summary file" ;
cwl:outputBinding [ CommandOutputBinding:glob
"$(inputs.input_bam.path.replace(/^.*[\\\\\\/]/, \"\").replace(/\\.[^/.]+$/, \"\") + \".spp_cross_corr.txt\")" ] ;
sld:type cwl:File .
rdfs:comment "-npeak=, threshold on number of peaks to call" ;
cwl:inputBinding [ CommandLineBinding:prefix "-npeak=" ;
CommandLineBinding:separate false
] ;
sld:type xsd:int , sld:null .
a cwl:CommandLineTool ;
rdfs:comment "\"bedToBigBed v. 2.7 - Convert bed file to bigBed. (BigBed version: 4)\nusage:\n bedToBigBed in.bed chrom.sizes out.bb\nWhere in.bed is in one of the ascii bed formats, but not including track lines\nand chrom.sizes is two column: \nand out.bb is the output indexed big bed file.\nUse the script: fetchChromSizes to obtain the actual chrom.sizes information\nfrom UCSC, please do not make up a chrom sizes from your own information.\nThe in.bed file must be sorted by chromosome,start,\n to sort a bed file, use the unix sort command:\n sort -k1,1 -k2,2n unsorted.bed > sorted.bed\"\n" ;
ns1:original_cwlVersion "v1.0" ;
cwl:arguments ( [ CommandLineBinding:position 4 ;
cwl:valueFrom "$(inputs.bed.path.replace(/^.*[\\\\\\/]/, '') + inputs.output_suffix)"
]
) ;
cwl:baseCommand ( "bedToBigBed" ) ;
cwl:cwlVersion ;
cwl:hints [ a ;
LoadListingRequirement:loadListing
"deep_listing"
] ;
cwl:hints [ a ;
NetworkAccess:networkAccess true
] ;
cwl:hints [ a cwl:DockerRequirement ;
DockerRequirement:dockerPull "dleehr/docker-hubutils"
] ;
cwl:inputs , , , , , , , , ,