---
tags: GeneLab
title: Baseline steps for running human-read removal
---

# Baseline steps for running human-read removal

## Creating environment

mamba create -n kraken2 -c conda-forge -c bioconda -c defaults kraken2==2.1.1 

Or one with nextflow also:

mamba create -n kraken2-nextflow -c conda-forge -c bioconda -c defaults kraken2==2.1.1 nextflow

---

## Setting up reference db

Run these wherever you want to keep the reference db (it's only about 3 GB):

# here is where I'm putting mine (will need to match what is in the code below)
mkdir -p ~/temp-kraken2-db
cd ~/temp-kraken2-db

# downloading
curl -L -o kraken2-human-db.tar.gz https://ndownloader.figshare.com/files/25627058

# unpacking and decompressing, then removing the tar
tar -xzvf kraken2-human-db.tar.gz && rm kraken2-human-db.tar.gz

---

## Getting tiny example data

These are less than a kB.

curl -L -o Sample-1_R1.fastq.gz https://figshare.com/ndownloader/files/43613283
curl -L -o Sample-1_R2.fastq.gz https://figshare.com/ndownloader/files/43613286

---

## Running it

conda activate kraken2

kraken2 --db ~/temp-kraken2-db/kraken2-human-db --gzip-compressed \
        --threads 2 --use-names --paired \
        --output Sample-1-kraken2-output.txt \
        --report Sample-1-kraken2-report.tsv \
        --unclassified-out "Sample-1_R#.fastq" \
        Sample-1_R1.fastq.gz Sample-1_R2.fastq.gz