Skip to content

Evaluation of long-read support for the MGnify pipeline

License

Notifications You must be signed in to change notification settings

hoelzer/mgnify-lr

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MGnify-LR Workflow

Maintainer: Martin Hölzer

Email: hoelzer.martin@gmail.com

Discontinued and not maintained

Input examples

  • one .fastq file per sample: --nano 'sample1.fastq'
  • paired end illumina: --illumina 'S_41_17_Cf*.R{1,2}.fastq.gz'

Hybrid data needs to have matching simpleNames such as sample1.fastq.gz (ONT) and sample1.R{1,2}.fastq.gz (Illumina).

Execution example

After pulling this git:

nextflow run main.nf --nano test_data/ERR3662306_1_small.fastq.gz -profile local,docker

or let nextflow pull this git:

nextflow run hoelzer/mgnify-lr --nano ~/.nextflow/assets/hoelzer/mgnify-lr/test_data/ERR3662306_1_small.fastq.gz --output ~/mgnify-lr_output -profile local,docker

to also decontaminate your reads and final assembly just add a --species like this:

nextflow run hoelzer/mgnify-lr --nano ~/.nextflow/assets/hoelzer/mgnify-lr/test_data/ERR3662306_1_small.fastq.gz --output ~/mgnify-lr_output --species eco -profile local,docker

Currently supported species with auto-download are:

  • hsa [Ensembl: Homo_sapiens.GRCh38.dna.primary_assembly]
  • mmu [Ensembl: Mus_musculus.GRCm38.dna.primary_assembly]
  • eco [Ensembl: Escherichia_coli_k_12.ASM80076v1.dna.toplevel]

Per default Illumina data will be cleaned for phiX phage and Nanopore data for DNA CS (DCS) spike-in. You can deactivate this via the --phix and --dcs flags.

Profiles

This workflow comes with some pre-defined profiles for local execution and execution on a HPC using Conda, Docker, and Singularity. You can combine different profiles to run the pipeline on different systems (local, LSF, SLURM, Cloud) using different technologies (Conda, Docker, Singularity).

For the most basic execution on a local machine using Docker use:

-profile local,docker

If you want to use Conda instead of Docker:

-profile local,conda

If you want to run on a HPC w/ LSF and Singularity use:

-profile lsf,singularity
# In that case you should also adjust --singularityCacheDir, --workdir, --databases to match your HPC environment

Workflow

Nanopore-only w/ decontamination

chart

Hybrid w/ metaSPAdes and decontamination

chart

Hybrid w/ flye and decontamination

chart

Execution examples of possible input and decontamination combinations

Below are execution examples for the test data in this repository. For each command ideel plots show the difference between the raw and polished assembly. Please keep in mind that this is only a small toy sample.

Nanopore-only

nextflow run main.nf --list --nano test_data/ont.csv --workdir work -profile local,docker --output results/test_ont-only

Nanopore-only w/ decontamination of ONT spike-in control

nextflow run main.nf --list --nano test_data/ont.csv --workdir work -profile local,docker -resume --output results/test_ont-only-clean --index_ont clean/ont/DCS.mmi --index_fna clean/assembly/DCS_FNA.mmi

Hybrid using SPAdes

nextflow run main.nf --list --nano test_data/ont.csv --illumina test_data/ill.csv --workdir work -profile local,docker -resume --output results/test_hybrid-spades 

Hybrid using SPAdes w/ decontamination of ONT and Illumina spike-in controls

nextflow run main.nf --list --nano test_data/ont.csv --illumina test_data/ill.csv --workdir work -profile local,docker -resume --output results/test_hybrid-spades-clean --index_ont clean/ont/DCS.mmi --index_fna clean/assembly/NC_001422_DCS.mmi --bbduk clean/NC_001422.fna.gz 

Hybrid using Flye

nextflow run main.nf --list --nano test_data/ont.csv --illumina test_data/ill.csv --workdir work -profile local,docker -resume --output results/test_hybrid-flye --assemblerHybrid flye

Hybrid using Flye w/ decontamination of ONT and Illumina spike-in controls

nextflow run main.nf --list --nano test_data/ont.csv --illumina test_data/ill.csv --workdir work -profile local,docker -resume --output results/test_hybrid-flye-clean --index_ont clean/ont/DCS.mmi --index_fna clean/assembly/NC_001422_DCS.mmi --bbduk clean/NC_001422.fna.gz --assemblerHybrid flye

About

Evaluation of long-read support for the MGnify pipeline

Resources

License

Stars

Watchers

Forks

Packages

No packages published