fabianackle / nestlink-pipeline

Pipeline to generate accurate consenus sequences of NestLink libraries.

NestLink-pipeline

NestLink-pipeline is a pipeline for processing NestLink libraries sequenced by nanopore sequencing. Reads are binned according to their flycodes (UMIs). Accurate consensus sequences are calculated using Medaka. Variants are called with the pipeline, resulting in a flycode assignment table that links protein variants to their respective set of flycodes.

Warning

NestLink-pipeline is still in development. Certain library-specific strings are still hard-coded in main.nf and have to be edited before running the pipeline.

Requirements

• Conda (https://conda-forge.org/)
• Nextflow (Installation guide)
• mini_align (mini_align.sh placed in projectDir/bin/)
• Medaka (Note: Medaka is not yet integrated and must be run separately)

Running the pipeline

1. Clone the repository.
2. Place the basecalled sequencing data and the reference sequence into projectDir/data/.
3. Run the first workflow "prepare_data" of the pipeline: nextflow run main.nf -entry prepare_data
4. Generate the consenus sequences using medaka with the data from projectDir/medaka_input/, and place the Medaka output assembly.fasta into the folder projectDir/medaka_input/.
5. Run the second workflow "nestlink" of the pipeline: nextflow run main.nf -entry nestlink

Generating consensus sequences using Medaka

Example with CUDA and Singularity installed on Ubuntu 20.04.

singularity run --nv \
    --bind /home/ubuntu/calculation/consensus:/data --pwd /data \
    docker://ontresearch/medaka:latest medaka consensus \
    --batch 200 --threads 2 --model r1041_e82_400bps_sup_v5.0.0  \
    merged.sorted.bam results.contigs.hdf

singularity run --nv \
    --bind /home/ubuntu/calculation/consensus:/data --pwd /data \
    docker://ontresearch/medaka:latest medaka stitch \
    results.contigs.hdf reference_all.fasta assembly.fasta