repeatModeler2_nf

A Nextflow wrapper for RepeatModeler2.

Dependencies:

Singularity , (typically has to be installed by an admin / system administrator)
Miniconda , (can be installed by a user, see https://docs.conda.io/en/latest/miniconda.html)
Nextflow , (once you have conda installed: conda install -y nextflow)

If you have set the --genome parameter, and have access to Miniconda & Singularity, this pipeline should 'just work' and automatically pull the necessary RepeatModeler developer provided Docker containers, and install other software via conda.

See here for more background on the dfam/tetools container: https://github.com/Dfam-consortium/TETools

Running the pipeline

git clone https://github.com/photocyte/repeatModeler2_nf.git
#After manually git cloning' the nextflow main.nf & nextflow.config into your working directory
nextflow run main.nf -resume -profile singularity --cpuNum 10 --genome examples/U00096.3.fasta
##-profile docker is also possible

#Letting nextflow manage the git cloning
nextflow pull https://github.com/photocyte/repeatModeler2_nf
nextflow run repeatModeler2_nf -latest -resume -profile singularity --cpuNum 10 --genome examples/U00096.3.fasta 
##-profile docker is also possible

Local vs cluster/HPC execution

By default, the workflow runs locally. If you'd instead prefer Nextflow to submit the processes/jobs to a high-performance-computing (HPC) cluster, edit the first line of nextflow.config, to specify your desired cluster type. See https://www.nextflow.io/docs/latest/executor.html for more details. I.e., for running on a PBS cluster change: params.executor = "local" to params.executor = "pbs"

Results

Look in ./results once the pipeline is complete

RepeatModeler2 citation

Flynn JM, Hubley R, Goubert C, Rosen J, Clark AG, Feschotte C, Smit AF. 2020. RepeatModeler2 for automated genomic discovery of transposable element families. PNAS. doi:10.1073/pnas.1921046117

Directed acyclic graph of pipeline execution

(Note, DAG rendering is a little broken currently)