Flashlite-Trinity contains two workflows that run Trinity on the University of Queensland's HPC, Flashlite. Trinity performs de novo transcriptome assembly of RNA-seq data by combining three independent software modules Inchworm, Chrysalis and Butterfly to process RNA-seq reads. The algorithm can detect isoforms, handle paired-end reads, multiple insert sizes and strandedness. Users can run Flashlite-Trinity on single samples, or smaller samples requiring <500Gb ...

Description: Trinity @ NCI-Gadi contains a staged Trinity workflow that can be run on the National Computational Infrastructure’s (NCI) Gadi supercomputer. Trinity performs de novo transcriptome assembly of RNA-seq data by combining three independent software modules Inchworm, Chrysalis and Butterfly to process RNA-seq reads. The algorithm can detect isoforms, handle paired-end reads, multiple insert sizes and strandedness. ...

SLURM HPC Cromwell implementation of GATK4 germline variant calling pipeline

See the GATK website for more information on this toolset

Assumptions

• Using hg38 human reference genome build
• Running using HPC/SLURM scheduling. This repo was specifically tested on Pawsey Zeus machine, primarily running in the /scratch partition.
• Starting from short-read Illumina paired-end fastq files as input

Dependencies

The following versions have been ...

Local Cromwell implementation of GATK4 germline variant calling pipeline

See the GATK website for more information on this toolset

Assumptions

• Using hg38 human reference genome build
• Running 'locally' i.e. not using HPC/SLURM scheduling, or containers. This repo was specifically tested on Pawsey Nimbus 16 CPU, 64GB RAM virtual machine, primarily running in the /data volume storage partition.
• Starting from short-read Illumina paired-end fastq ...

A porting of the Trinity RNA assembly pipeline, https://trinityrnaseq.github.io, that uses Nextflow to handle the underlying sub-tasks. This enables additional capabilities to better use HPC resources, such as packing of tasks to fill up nodes and use of node-local disks to improve I/O. By design, the pipeline separates the workflow logic (main file) and the cluster-specific configuration (config files), improving portability.

Based on a pipeline by Sydney Informatics Hub: ...