bd2kgenomics / fusion

UCSC Treehouse Gene Fusion Analysis Pipeline

STAR-Fusion for Treehouse RNA-seq analysis

Overview

Gene fusions play a major role in tumorigenesis, so it is crucial that Treehouse has a pipeline for detecting them. We have built a docker container that runs STAR-Fusion and filters the output against a list of known cancer fusion genes. There is also an option to run additional filters and generate de novo assembled fusion transcripts using the FusionInspector program.

Docker and usage

Image located on hub.docker.com

REPOSITORY: ucsctreehouse/fusion

TAG: 0.1.0

IMAGE ID: 5eeb0e139cd6

Input files

The pipeline requires paired-end fastq files, the output directory, and the genome library directory. The genelist is already baked into the docker container, but there is an option to include a different genelist. Please refer to the STAR-Fusion documentation for creating a genome library. You can also find a prebuilt genome library here: http://ceph-gw-01.pod/references/STARFusion-GRCh38gencode23.tar.gz

Wraps STAR-Fusion program and filters output using FusionInspector.

optional arguments:
  -h, --help            show this help message and exit
  --left_fq R1          Fastq 1
  --right_fq R2         Fastq 2
  --output_dir OUTPUT_DIR
                        Output directory
  --genome_lib_dir GENOME_LIB_DIR
                        Reference genome directory
  --CPU CPU             Number of jobs to run in parallel
  --genelist GENELIST
  --skip-filter
  -F, --run_fusion_inspector
                        Runs FusionInspector on STAR-Fusion output
  --clean               Cleans output directory
  --test

Run command

docker run -it --rm -v `pwd`:/data ucsctreehouse/fusion:0.1.0 \
                                   --left_fq 1.fq.gz \
                                   --right_fq 2.fq.gz \
                                   --output_dir fusion_output \
                                   --CPU `nproc` \
                                   --genome_lib_dir STARFusion-GRCh38gencode23
                                   --run_fusion_inspector
                                   --clean

Output

There will be many files in the output directory, but you can find the fusion calls in the fusion-output directory:

• star-fusion.fusion_candidates.final.abridged.FFPM
• star-fusion.fusion_candidates.final.in_genelist.abridged.FFPM

The second file contains fusion calls where both fusion partners are in the gene-list. If the pipeline is run with the --run_fusion_inspector flag, then there will also be a separate FusionInspector directory that contains fusion calls that passed the FusionInspector filter. FusionInspector is also configured to de novo assemble fusion transcripts using Trinity and to create IGV input files for viewing fusion sequences.

• FusionInspector predictions
  • FusionInspector/FusionInspector.fusion_predictions.final.abridged.FFPM
• de novo assembled transcripts
  • FusionInspector/FusionInspector.gmap_trinity_GG.fusions.fasta
• IGV input files
  • FusionInspector/FusionInspector.fa
  • FusionInspector/FusionInspector.gtf
  • FusionInspector/FusionInspector.junction_reads.bam
  • FusionInspector/FusionInspector.spanning_reads.bam

FusionInspector predictions with a combined FFPM > 0.1 are considered significant.