Tool runs preseq lc_extrap. Only BAM input file is supported (-B option is used by default) successCodes: [1] - is used to pass this tool as a step in a workflow in case the BAM file was not correct for Preseq Discarded arguments: -V, -vals input is a text file containing only the observed counts -H, -hist input is a text file containing the observed histogram

Tool processes and combines log files generated by Trimgalore, Bowtie, Samtools and MACS2.

`get_output_prefix` function returns output file prefix equal to `output_prefix`+`_collected_statistics_report` (if this input is provided) or generated on the base of bowtie log basename with `_collected_statistics_report` extension.

Excludes chromosomes from the input BAM file. Filters reads by quality. If there is only one chromosome present, you cannot exclude it

Tool runs Trimgalore - the wrapper around Cutadapt and FastQC to consistently apply adapter and quality trimming to FastQ files.

`default_log_name` function returns names for generated log files (for both paired-end and single-end cases). `trim_galore` itself doesn't support setting custom names for output files.

For paired-end data processing both `input_file_pair` and `paired` should be set. If either of them is not set, the other one becomes unset automatically.

If input trigger was set to false, skip running trimaglore and return unchanged input files

Workflow converts input BAM file into bigWig and bedGraph files.

Input BAM file should be sorted by coordinates (required by `bam_to_bedgraph` step).

If `split` input is not provided use true by default. Default logic is implemented in `valueFrom` field of `split` input inside `bam_to_bedgraph` step to avoid possible bug in cwltool with setting default values for workflow inputs.

`scale` has higher priority over the `mapped_reads_number`. The last one is used to calculate `-scale` parameter for `bedtools genomecov` (step `bam_to_bedgraph`) only in a case when input `scale` is not provided. All logic is implemented inside `bedtools-genomecov.cwl`.

`bigwig_filename` defines the output name only for generated bigWig file. `bedgraph_filename` defines the output name for generated bedGraph file and can influence on generated bigWig filename in case when `bigwig_filename` is not provided.

All workflow inputs and outputs don't have `format` field to avoid format incompatibility errors when workflow is used as subworkflow.

Tool maps input raw reads files to reference genome using Bowtie.

`default_output_filename` function returns default name for SAM output and log files. In case when `sam` and `output_filename` inputs are not set, default filename will have `.sam` extension but format may not correspond SAM specification. To set output filename manually use `output_filename` input. Default output filename is based on `output_filename` or basename of `upstream_filelist`, `downstream_filelist` or `crossbow_filelist` file (if array, the first file in array is taken). If function is called without argenments and `output_filename` input is set, it will be returned from the function.

For single-end input data any of the `upstream_filelist` or `downstream_filelist` inputs can be used.

Log filename (`log_file` output) is generated by `default_output_filename` function with ex='.bw'

`indices_folder` defines folder to contain Bowtie indices. Based on the first found file with `rev.1.ebwt` or `rev.1.ebwtl` extension, bowtie index prefix is returned from input's `valueFrom` field.

Tool runs peak calling using MACS2 within the custom script `run.py`

`default_output_filename` function returns default output prefix for all generated output files. Function is called when `output_prefix` input is not set. Optional argument `sufix` is equal to `_macs2` by default. `sufix` is hardcoded and should not be changed (if necessary, update `_macs2` in all tool outputs and `run.py` script too). `sufix` is included in the filenames of all output files except those, who ends with `_fragment_stat.tsv`. For the rest of the output files `_macs2` is already included as a part of the `sufix` argument when `default_output_filename` function is called. This is mostly done because of the naming conventions for output files in supplemental project `https://github.com/Barski-lab/biowardrobe-analysis` (if `_fragment_stat.tsv` is updated to `_macs2_fragment_stat.tsv` in `biowardrobe-analysis`, the way we use `default_output_filename` function can be simplified). The basename for output filename is taken from `treatment_file` input (if array - the first item is used).

`baseCommand` runs Python script `run.py` staged into the output directory. Within this script MACS2 is called one ortwo times as a subprocess with all provided input parameters.

`extsize` is mandatory parameter. It will be ignored by MASC2, when `--nomodel` is not set.

The aim of `run.py` script is to return `calculated_fragment_size`, `expected_fragment_size` and `island_count` as a text file.

Scenarios: 1. Run MACS2 with all input parameters. If MACS2 first run failed and `--nomodel` wasn't set (it means that `--extsize` wasn't used), set `expected_fragment_size` to `--extsize` value and try to rerun MACS2 with `--nomodel` parameter. If MACS2 second run failed too - exit workflow, if it didn't fail - get values for `calculated_fragment_size` and `island_count` from generated output file. Returned data: `calculated_fragment_size` - the value of fragment size from MACS2 second run `expected_fragment_size` - the value of fragment size equal to `--extsize` (set when MACS2 first run failed) `island_count` - the number of islands from MACS2 second run

2. Run MACS2 with all input parameters. If MACS2 first run failed and `--nomodel` was already set, exit workflow, because there is no need to rerun MACS2, it will fail too.

3. Run MACS2 with all input parameters. If MACS2 first run didn't fail, we get `calculated_fragment_size` and `island_count` from generated output file, and set `expected_fragment_size` to be equal to `calculated_fragment_size`. If `calculated_fragment_size` is bigger or equal to 80, there is no need to rerun MACS2. Returned data: `calculated_fragment_size` - the value of fragment size from MACS2 first run `expected_fragment_size` - the value of fragment size equal to `calculated_fragment_size` `island_count` - the number of islands from MACS2 first run

4. Run MACS2 with all input parameters. If MACS2 first run didn't fail, we get `calculated_fragment_size` and `island_count` from generated output file, and set `expected_fragment_size` to be equal to `calculated_fragment_size`. If `calculated_fragment_size` is less then 80 and MACS2 was run without '--nomodel' (it means we didn't force it to use the fixed fragment size `--extsize`), we rerun MACS2 with `--nomodel` argument. If MACS2 second run failed too - exit workflow, if not - get values for `calculated_fragment_size` and `island_count` from generated output file. Returned data: `calculated_fragment_size` - the value of fragment size from MACS2 second run `expected_fragment_size` - the value of fragment size from MACS2 first run `island_count` - the number of islands from MACS2 second run

5. Run MACS2 with all input parameters. If MACS2 first run didn't fail and was already run with '--nomodel' parameter (forced to use fixed fragment size `--extsize`), we get `calculated_fragment_size` and `island_count` from generated output file, and set `expected_fragment_size` to be equal to `calculated_fragment_size`. No mater which value we got for `calculated_fragment_size`, we don't rerun MACS2. Returned data: `calculated_fragment_size` - the value of fragment size got from MACS2 first run `expected_fragment_size` - the value of fragment size equal to `calculated_fragment_size` `island_count` - the number of islands got from MACS2 first run of MACS2

Tool to remove duplicates from coordinate sorted BAM file set as input `bam_file`. If input `trigger` is set to `true` or isn't set at all (`true` is used by default), run `samtools rmdup`, return newly generated BAM file without duplicates and log as outputs `rmdup_output` and `rmdup_log`. If input `trigger` is set to `false`, return unchanged BAM and index files (if provided in secondaryFiles), previously staged into output directory, and log.

Before execution `baseCommand` input BAM and index files (if provided in secondaryFiles) are staged into directory set as docker parameter `--workdir` (tool's output directory), using `InitialWorkDirRequirement`. Setting `writable: true` makes cwl-runner to make copies of the BAM and index files (if provided in secondaryFiles) and mount them to docker container with `rw` mode as part of `--workdir` (if set to false, the files staged into output directory will be mounted to docker container separately with `ro` mode). Because `bam_file` is copied into `--workdir` (where `samtools rmdup` should be run) we need to change its name - set `entryname`. This prevents `samtools rmdup` from reading and writing to the same file (which happens only when `output_filename` is not set and we generated output filename automatically on the base of the `bam_file` basename). Index file which is optionally set as `secondaryFiles` for `bam_file` shouln't be renamed, because `samtools rmdup` don't work with this file at all and is completely ignored when `trigger` is true.

Trigger logic is implemented in bash script set by default in input `bash_script`. If first argment $0 (which is `trigger` input) is true, run `samtools rmdup` with the rest of the arguments. If $0 is not true, skip samtools running and rename staged into output directory BAM file. Current filename of staged BAM file and the filename which we are waiting to be the output of tool are always set as two last arguments of the script.

Input `trigger` is Boolean, but returns String, because of `valueFrom` field. The `valueFrom` is used, because if `trigger` is false, cwl-runner doesn't append this argument at all to the the `baseCommand` - new feature of CWL v1.0.2. Alternatively, `prefix` field could be used, but it causes changing logic in bash script saved in `bash_script` input.

`default_output_filename` function is used for generating output filename if input `output_filename` is not set or in case when `trigger` is false and we need to return BAM and index files (if provided in secondaryFiles) staged into output directory.

Output `rmdup_output` returns `secondaryFiles` only in case when `trigger` was set to false (we need to rerun index).

Tool renames `source_file` to `target_filename`. Input `target_filename` should be set as string. If it's a full path, only basename will be used. If BAI file is present, it will be renamed too

Tool assigns each peak obtained from MACS2 to a gene and region (upstream, promoter, exon, intron, intergenic)

`default_output_filename` function returns output filename with sufix set as `ext` argument. Function is called when either `output_filename` or `log_filename` inputs are not provided.

Tool renames `source_file` to `target_filename`. Input `target_filename` should be set as string. If it's a full path, only basename will be used. If BAI file is present, it will be renamed too

Generates statistics for the input BAM file. Output if filtered with grep SN | cut -f 2-

Tool calculates average tag density profile around all annotated TSS.

`default_output_filename` function returns output filename with sufix set as `ext` argument. Function is called when either `output_filename` or `log_filename` inputs are not provided.

Before running `baseCommand`, annotaion file `annotation_filename` is staged into output directory (Docker's `--workdir`) with `\"writable\": true` (to allow to overwrite it by `refgene-sort`).

To run `atdp` index file should be provided (either in `secondaryFiles` of `input_file` or as separate input `index_file`)

`baseCommand` runs bash script from `script` input. Script runs `refgene-sort` to sort annotaion file and then runs `atdp`. `refgene-sort` - utility to sort refgene annotation files, using MySQL syntax.

Optionally (with cwltool==1.0.20171107133715), script can be simplified to #!/bin/bash set -- \"$0\" \"$@\" refgene-sort -i \"${2:4}\" -o \"${2:4}\" -s \"ORDER BY chrom, strand, CASE strand WHEN '+' THEN txStart WHEN '-' THEN txEnd END\" atdp \"$@\" because `set -- \"$1\" --a=$(basename \"${2:4}\") \"${@:3}\"` is not needed anymore.

Tool to sort and index input BAM/SAM/CRAM. If input `trigger` is set to `true` or isn't set at all (`true` is used by default), run `samtools sort` and `samtools index`, return sorted BAM and BAI/CSI index file. If input `trigger` is set to `false`, return unchanged `sort_input` (BAM/SAM/CRAM) and index (BAI/CSI, if provided in `secondaryFiles`) files, previously staged into output directory.

Before execution `baseCommand`, `sort_input` and `secondaryFiles` (if provided) are staged into directory set as docker parameter `--workdir` (tool's output directory), using `InitialWorkDirRequirement`. Setting `writable: true` makes cwl-runner to make copies of the `sort_input` and `secondaryFiles` (if provided) and mount them to docker container with `rw` mode as part of `--workdir` (if set to false, the files staged into output directory will be mounted to docker container separately with `ro` mode). Because both `samtools sort` and `samtools index` can overwrite files with the same names (and in case of `samtools sort` even the input file can be overwritten), we don't need to rename any of the staged files.

Trigger logic is implemented in two bash scripts set by default as `bash_script_sort` and `bash_script_index` inputs. For both of then, if the first argument $0 (which is `trigger` input) is true, run `samtools sort/index` with the rest of the arguments. If $0 is not true, skip `samtools sort/index` and return `sort_input` and `secondaryFiles` (if provided) staged into output directory.

Input `trigger` is Boolean, but returns String, because of `valueFrom` field. The `valueFrom` is used, because if `trigger` is false, cwl-runner doesn't append this argument at all to the the `baseCommand` - new feature of CWL v1.0.2. Alternatively, `prefix` field could be used, but it causes changing in script logic.

If using `sort_output_filename`, the output file extension should be `*.bam`, because `samtools sort` defines the output file format on the base of the file extension. If `*.sam` is sed as output filename, it cannot be usefully indexed by `samtools index`.

`default_bam` function is used to generate output filename for `samtools sort` if input `sort_output_filename` is not set or when `trigger` is false and we need to return `sort_input` and `secondaryFiles` (if provided) files staged into output directory. Output filename is generated on the base of `sort_input` basename with `.bam` extension by default.

`ext` function is used to return the index file extension (BAI/CSI) based on `csi` and `bai` inputs according to the following logic `csi` && `bai` => BAI !`csi` && !`bai ` => BAI `csi` && !`bai ` => CSI

Tool to decompress input FASTQ file Bash script's logic: - disable case sensitive glob check - check if root name of input file already include '.fastq' or '.fq' extension. If yes, set DEFAULT_EXT to \"\" - check file type, decompress if needed - return 1, if file type is not recognized This script also works of input file doesn't have any extension at all

Tool to decompress input FASTQ file Bash script's logic: - disable case sensitive glob check - check if root name of input file already include '.fastq' or '.fq' extension. If yes, set DEFAULT_EXT to \"\" - check file type, decompress if needed - return 1, if file type is not recognized This script also works of input file doesn't have any extension at all

Tool calculates statistics on the base of FASTQ file quality scores. If `output_filename` is not provided call function `default_output_filename` to return default output file name generated as `input_file` basename + `.fastxstat` extension.

Tool calculates statistics on the base of FASTQ file quality scores. If `output_filename` is not provided call function `default_output_filename` to return default output file name generated as `input_file` basename + `.fastxstat` extension.

Tool to sort and index input BAM/SAM/CRAM. If input `trigger` is set to `true` or isn't set at all (`true` is used by default), run `samtools sort` and `samtools index`, return sorted BAM and BAI/CSI index file. If input `trigger` is set to `false`, return unchanged `sort_input` (BAM/SAM/CRAM) and index (BAI/CSI, if provided in `secondaryFiles`) files, previously staged into output directory.

Before execution `baseCommand`, `sort_input` and `secondaryFiles` (if provided) are staged into directory set as docker parameter `--workdir` (tool's output directory), using `InitialWorkDirRequirement`. Setting `writable: true` makes cwl-runner to make copies of the `sort_input` and `secondaryFiles` (if provided) and mount them to docker container with `rw` mode as part of `--workdir` (if set to false, the files staged into output directory will be mounted to docker container separately with `ro` mode). Because both `samtools sort` and `samtools index` can overwrite files with the same names (and in case of `samtools sort` even the input file can be overwritten), we don't need to rename any of the staged files.

Trigger logic is implemented in two bash scripts set by default as `bash_script_sort` and `bash_script_index` inputs. For both of then, if the first argument $0 (which is `trigger` input) is true, run `samtools sort/index` with the rest of the arguments. If $0 is not true, skip `samtools sort/index` and return `sort_input` and `secondaryFiles` (if provided) staged into output directory.

Input `trigger` is Boolean, but returns String, because of `valueFrom` field. The `valueFrom` is used, because if `trigger` is false, cwl-runner doesn't append this argument at all to the the `baseCommand` - new feature of CWL v1.0.2. Alternatively, `prefix` field could be used, but it causes changing in script logic.

If using `sort_output_filename`, the output file extension should be `*.bam`, because `samtools sort` defines the output file format on the base of the file extension. If `*.sam` is sed as output filename, it cannot be usefully indexed by `samtools index`.

`default_bam` function is used to generate output filename for `samtools sort` if input `sort_output_filename` is not set or when `trigger` is false and we need to return `sort_input` and `secondaryFiles` (if provided) files staged into output directory. Output filename is generated on the base of `sort_input` basename with `.bam` extension by default.

`ext` function is used to return the index file extension (BAI/CSI) based on `csi` and `bai` inputs according to the following logic `csi` && `bai` => BAI !`csi` && !`bai ` => BAI `csi` && !`bai ` => CSI

Generates statistics for the input BAM file. Output if filtered with grep SN | cut -f 2-