Skip to content

edgardomortiz/NanoAmp

Repository files navigation

NanoAmp

Species identification via Nanopore sequencing of barcoded amplicons

Installation:

NanoAmp was tested on Linux, but should work on Mac as well. The dependencies can be installed in a separate conda environment with the following command:

conda create -n nanoamp -c bioconda -c conda-forge canu=2.2 parallel medaka=1.6.0 longshot tabix perl-vcftools-vcf bbmap ucsc-blat
conda activate nanoamp

Afterwards, this repository must be cloned:

git clone https://github.com/edgardomortiz/NanoAmp

Finally, add the directory where you cloned NanoAmp to your system $PATH

Usage:

All steps can be run with a single command:

nanoamp.sh RUN_NAME CORES

Where RUN_NAME is the name of the sequencing run, assigned in MinKNOW. CORES indicates the number of processors to use.

However, if you wan to run steps manually, the order of execution is:

  1. Basecalling: basecall_demultiplex_guppy.sh
  2. Quality filtering: filter_quality_bbduk.sh
  3. Assembly: assemble_canu.sh
  4. Polish assembly: polish_raw_medaka.sh
  5. Separate haplotypes: haplotype_polished_longshot.sh
  6. Identify samples: match_db.py

For help on each script just execute the script without extra arguments.

About

Species identification via Nanopore sequencing of barcoded amplicons

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published