rmcolq/rampart-coi

RAMPART protocol for classifier

rampart-coi

RAMPART protocol for mitochondrial COI based classifier.

Installing and set up

1. First clone the repository

   git clone https://github.com/rmcolq/rampart-coi.git
   cd rampart-coi
   

2. Setup and activate the conda environment

   conda env create -f environment.yml
   source activate rampart-coi
   

3. Download (large) reference files and build kraken database (takes about 10 minutes)

   First make a kraken data directory (mkdir kraken_db/data) and copy two big files

   • BOLD-Animalia-COI5p_old.fasta
   • NCBI_Animalia_BOLD_COI_250819.fasta from dropbox (VirusEvolution Dropbox/Group/Seawater_WIMP/kraken_data/) to kraken_db/data

   Then construct the database with:

   cd kraken_db/
   bash create_db.sh 
   cd ..
   

4. (Optional) Copy references file BOLD_COI_references_by_phylum.fasta.gz from dropbox (VirusEvolution Dropbox/Group/Seawater_WIMP/heatmap_references/) and rename as references.fasta.

   Alternatively, download relevant references from NCBI by searching the nucleotide database for [species of interest] coi and filtering search results (Species=Animals, Compartment=Mitochondrion, Sequence Type=Nucleotide). Download by clicking on the send to drop down option at the top of the search results (Choose Destination=File, Format=FASTA, Sort by=Taxonomy ID, uncheck Show GI). Then run a python script to get this references file in the correct format using:

   python3 process_reference.py [downloaded_fasta]
   

Running

Create run folder:

mkdir [run_name]
cd [run_name]

Where [run_name] is whatever you are calling todays run (as specified in MinKNOW).

Create a basic run_configuration.json file in this run directory, e.g.

{
  "title": "My metagenomic run",
  "basecalledPath": "~/MinKNOW/data/reads/[run_name]/pass",
  "samples": [
    {
      "name": "Sample1",
      "description": "",
      "barcodes": [ "NB01" ]
    }
  ]
}

Run RAMPART from this run directory:

rampart --protocol /path/to/rampart-coi

Open a web browser to view http://localhost:3000 the RAMPART run.

Open /path/to/[run_name]/annotations/classified/all_krona.html in a second browser tab. Refresh this page periodically to view the latest krona results as they come in. NB this file does not exist until the pipeline is running.

Example dataset

cd example_data/millport
rampart
  --protocol ../../../rampart-coi
  --basecalledPath basecalled/pass 
  --referencesPath ../../BOLD_COI_references_by_phylum.fasta.gz 

NB you need to run this from your run directory and have a run_configuration.json file in this local directory. To view and explore the Krona plot, open annotations/classified/all_krona.html in your browser.

You can assemble the classified sequences where more than 100 reads have mapped to a species, e.g.

snakemake 
  --snakefile /path/to/rampart-coi/pipelines/assemble_found_taxa/Snakefile 
  --configfile /path/to/rampart-coi/pipelines/assemble_found_taxa/config.yaml 
  --config output_path=annotations/ 
  kraken_fasta=/path/to/rampart-coi/kraken_db/data/BOLD-Animalia-COI5p_old.fasta