ww-salmon RNA-seq Workflow

Overview

This workflow performs RNA-seq quantification using Salmon, a fast and accurate tool for transcript expression estimation. The workflow is designed to be simple to use while implementing best practices for RNA-seq analysis.

What this Workflow Does

1. Builds a Salmon Index from your reference transcriptome
2. Quantifies Transcripts for each of your RNA-seq samples
3. Generates Expression Matrices combining results from all samples

Requirements

• Cromwell or another WDL-compatible workflow engine
• Docker (the workflow uses the combinelab/salmon container)
• Input files:
  • Reference transcriptome (FASTA format)
  • RNA-seq reads (FASTQ format, can be gzipped)

Quick Start

1. Download the WDL file from this repository:

2. Create an inputs JSON file (e.g., inputs.json):

   {
     "SalmonRnaSeq.transcriptome_fasta": "path/to/transcriptome.fa",
     "SalmonRnaSeq.fastq_r1_files": [
       "path/to/sample1_R1.fastq.gz",
       "path/to/sample2_R1.fastq.gz"
     ],
     "SalmonRnaSeq.fastq_r2_files": [
       "path/to/sample1_R2.fastq.gz",
       "path/to/sample2_R2.fastq.gz"
     ]
   }

3. Run the workflow with Cromwell:

   java -jar cromwell.jar run salmon_rnaseq.wdl -i inputs.json
   

Input Parameters

Parameter	Description	Required?
transcriptome_fasta	Reference transcriptome in FASTA format	Yes
fastq_r1_files	Array of FASTQ files for read 1 (or single-end reads)	Yes
fastq_r2_files	Array of FASTQ files for read 2 (for paired-end data)	No
salmon_docker	Docker image for Salmon (default: "combinelab/salmon:latest")	No

Outputs

Output	Description
salmon_index_tar	Compressed Salmon index (can be reused for future analyses)
salmon_quant_dirs	Compressed quantification results for each sample
merged_tpm_matrix	Combined TPM values matrix for all samples
merged_counts_matrix	Combined read counts matrix for all samples

Examples

For Paired-End Data

{
  "SalmonRnaSeq.transcriptome_fasta": "references/gencode.v38.transcripts.fa",
  "SalmonRnaSeq.fastq_r1_files": [
    "samples/sample1_R1.fastq.gz",
    "samples/sample2_R1.fastq.gz"
  ],
  "SalmonRnaSeq.fastq_r2_files": [
    "samples/sample1_R2.fastq.gz", 
    "samples/sample2_R2.fastq.gz"
  ]
}

For Single-End Data

{
  "SalmonRnaSeq.transcriptome_fasta": "references/gencode.v38.transcripts.fa",
  "SalmonRnaSeq.fastq_r1_files": [
    "samples/sample1.fastq.gz",
    "samples/sample2.fastq.gz"
  ]
}

Common Questions

How do I get a transcriptome file?

You can download reference transcriptomes from:

• GENCODE (human/mouse)
• Ensembl (many species)
• UCSC Genome Browser

For human, a common choice is the GENCODE reference:

wget https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_38/gencode.v38.transcripts.fa.gz
gunzip gencode.v38.transcripts.fa.gz

How do I extract the results?

The workflow provides compressed output directories for each sample. To extract a specific sample's results:

tar -xzf sample1_quant.tar.gz

This will create a directory sample1_quant containing Salmon's output files, including:

• quant.sf: The main quantification results file
• logs/: Directory containing Salmon log files
• lib_format_counts.json: Information about the library type

What are TPM and counts?

• TPM (Transcripts Per Million): Normalized expression values suitable for comparing expression levels between samples
• counts: Estimated number of fragments/reads from each transcript, suitable for differential expression analysis

Under the Hood

This workflow:

1. Creates a Salmon index from your transcriptome
2. Processes each sample with optimal settings:
   • Automatic library type detection
   • GC bias correction
   • Sequence-specific bias correction
   • Mapping validation
3. Combines results into unified matrices with properly labeled sample names

Troubleshooting

Error: "Docker image not found"

• Solution: Ensure Docker is installed and running

Error: "File not found"

• Solution: Check the paths in your inputs.json file

Error: "Memory allocation failed"

• Solution: Adjust the memory_gb parameters in the WDL file

Advanced Customization

If you need to modify the workflow for advanced settings:

1. Edit the runtime parameters at the task level:

   runtime {
       docker: docker_image
       memory: "~{memory_gb} GB"
       cpu: cpu
       disks: "local-disk ~{disk_size_gb} SSD"
       preemptible: 1
   }

2. Add additional Salmon parameters in the command sections if needed

Need Additional Help?

• Salmon Documentation
• WDL Documentation
• Cromwell Documentation