GDC RNA-Seq Alignment Workflow

This workflow takes a set of input RNA-Seq short read data as FASTQ or BAM files and generates multiple harmonized BAM files, gene counts, and other datasets.

External Users

The entrypoint CWL workflow for external users is rnaseq-star-align/subworkflows/gdc_rnaseq_main_workflow.cwl.

The example input json in rnaseq-star-align/example/main_workflow.example.input.json.

Inputs

Name	Type	Description
readgroup_bam_file_list	readgroup_bam_file[]	array of objects containing BAM files and readgroup metadata
readgroup_fastq_file_list	readgroup_fastq_file[]	array of objects containing FASTQ files and readgroup metadata
ref_flat	File	ref flat annotation file
ribosome_intervals	File	interval file containing rRNA locations
star_genome_dir	Directory	the directory containing the STAR index files
gene_info	File	tab-separated file relating gene symbol, biotype, and other info to gene ID
threads	int?	the number of threads to use for multi-threaded tools
job_uuid	string	string used as a prefix for all the output filenames
picard_java_mem	int	amount of memory (Gb) to use for picard (default: 4)
gencode_version	string	string indicating gencode annotation version (default: v36)

Custom Data Types

• readgroup_bam_file - contains a bam file and an array of readgroup_meta objects

Name	Type	Description
bam	File	input aligned or unaligned bam file
readgroup_meta_list	readgroup_meta[]	array of readgroup_meta objects

• readgroup_meta - contains readgroup tags and values

Name	Type	Description
CN	string?	optional sequencing center
DS	string?	optional description
DT	string?	optional ISO8601 sequencing date
FO	string?	optional flow order array of nocleotide bases that corresponded to the nucleotides used for each flow of each read
ID	string	required read group ID
KS	string?	optional array of nucleotide bases that correspond to the key sequence of each read
LB	string?	optional library ID
PI	string?	optional predicted median insert size
PL	string	required platform
PM	string?	optional platform model
PU	string?	optional platform unit
SM	string	required sample ID

• readgroup_fastq_file - contains single or pair of fastq files and an array of readgroup_meta objects

Name	Type	Description
forward_fastq	File	read1 fastq file
reverse_fastq	File?	optional read2 fastq file if paired library
readgroup_meta_list	readgroup_meta[]	array of readgroup_meta objects

Outputs

Name	Type	Description
out_metrics_db	File	sqlite file containing metrics data
out_gene_counts_file	File	gene-level counts as estimated by STAR
out_junctions_file	File	TSV containing splice junctions detected by STAR
out_transcriptome_bam_file	File?	If there are paired-end reads, the transcriptome alignments are provided in this unsorted bam file
out_chimeric_bam_file	File?	If there are paired-end reads, the chimeric alignments (sorted and indexed)
out_chimeric_tsv_file	File?	If there are paired-end reads, the TSV containing chimeric information for fusion detection
out_genome_bam	File	the final genome aligned bam (sorted and indexed)
out_archive_file	File	tar.gz archive containing other outputs from STAR

GDC Users

The entrypoint CWL workflow for GDC users is rnaseq-star-align/star2pass.rnaseq_harmonization.cwl. Additionally as a special case can handle a limited set of tar files.