Dockerfile

FROM ubuntu:xenial

# for easy upgrade later. ARG variables only persist during build time.
ARG K2VER="2.1.0"

LABEL base.image="ubuntu:xenial"
LABEL dockerfile.version="2"
LABEL software="Kraken2"
LABEL software.version="2.1.0"
LABEL description="Taxonomic sequence classifier"
LABEL website="https://github.com/DerrickWood/kraken2"
LABEL license="https://github.com/DerrickWood/kraken2/blob/master/LICENSE"
LABEL maintainer="Curtis Kapsak"
LABEL maintainer.email="pjx8@cdc.gov"

# install dependencies and cleanup apt garbage
RUN apt-get update && apt-get -y --no-install-recommends install \
 wget \
 ca-certificates \
 zlib1g-dev \
 make \
 g++ \
 rsync \
 cpanminus && \
 rm -rf /var/lib/apt/lists/* && apt-get autoclean

# perl module required for kraken2-build
RUN cpanm Getopt::Std

# DL Kraken2, unpack, and install
RUN wget https://github.com/DerrickWood/kraken2/archive/v${K2VER}.tar.gz && \
 tar -xzf v${K2VER}.tar.gz && \
 rm -rf v${K2VER}.tar.gz && \
 cd kraken2-${K2VER} && \
 ./install_kraken2.sh . && \
 mkdir /data /kraken2-db

ENV PATH="$PATH:/kraken2-${K2VER}" \
    LC_ALL=C

WORKDIR /data

##### NO DATABASE INCLUDED WITH THIS DOCKER IMAGE #####
## User will need to mount a directory from their host machine that contains kraken2 database files 
## to a directory in the container (/kraken2-db exists for this purpose, but feel free to use another location)


# DL MiniKraken2_8GB database. Built from RefSeq bacteria, archaea, viral, and human libraries.
# --strip-components=1 used so that the *.k2d files end up inside /kraken2-db and not another directory
#RUN mkdir /kraken2-db && \
# cd /kraken2-db && \
# wget --no-check-certificate https://genome-idx.s3.amazonaws.com/kraken/minikraken2_v2_8GB_201904.tgz && \
# tar -zxf --strip-components=1 minikraken2_v2_8GB_201904.tgz && \
# rm -rf minikraken2_v2_8GB_201904.tgz