Dockerfile

# Stage 1: Build stage
ARG BINDASHTREE_VER="0.1.1"

FROM ubuntu:jammy AS builder

# Set global variables
ARG BINDASHTREE_VER

# Update package manager and install necessary tools
RUN apt-get update && apt-get install -y --no-install-recommends \
    wget \
    curl \
    build-essential \
    gcc \
    pkg-config \
    libssl-dev \
    ca-certificates \
    && apt-get clean && rm -rf /var/lib/apt/lists/*

# Install Rust and Cargo using rustup
RUN curl https://sh.rustup.rs -sSf | sh -s -- -y && \
    export PATH="$HOME/.cargo/bin:$PATH" && \
    rustup default stable

# Ensure Rust and Cargo are available
ENV PATH="/root/.cargo/bin:$PATH"

# Download, extract, and build bindashtree
RUN wget -q https://github.com/jianshu93/bindashtree/archive/refs/tags/v${BINDASHTREE_VER}.tar.gz && \
    tar -xzvf v${BINDASHTREE_VER}.tar.gz && \
    cd bindashtree-${BINDASHTREE_VER} && \
    /root/.cargo/bin/cargo build --release

FROM ubuntu:jammy AS app
ARG BINDASHTREE_VER

# Install wget for test stage compatibility
RUN apt-get update && apt-get install -y --no-install-recommends \
    wget \
    ca-certificates \
    && apt-get clean && rm -rf /var/lib/apt/lists/*

# Labels for metadata
LABEL base.image="ubuntu:jammy" 
LABEL dockerfile.version="1" 
LABEL software="bindashtree" 
LABEL software.version="${BINDASHTREE_VER}" 
LABEL description="Binwise Densified MinHash and Rapid Neighbor-joining Tree Construction for microbial genomes." 
LABEL website="https://github.com/jianshu93/bindashtree" 
LABEL license.url="https://github.com/jianshu93/bindashtree?tab=MIT-1-ov-file#readme" 
LABEL maintainer="Taylor K. Paisie" 
LABEL maintainer.email="ltj8@cdc.gov"

# Copy built binaries from the builder stage
COPY --from=builder /bindashtree-${BINDASHTREE_VER}/target/release/bindashtree /usr/local/bin/

CMD ["bindashtree", "--help"]

WORKDIR /data

FROM app AS test

WORKDIR /data/test

# Download test files
RUN wget -q https://ftp.ncbi.nlm.nih.gov/genomes/all/GCA/002/587/385/GCA_002587385.1_ASM258738v1/GCA_002587385.1_ASM258738v1_genomic.fna.gz && \
    wget -q https://ftp.ncbi.nlm.nih.gov/genomes/all/GCA/002/596/765/GCA_002596765.1_ASM259676v1/GCA_002596765.1_ASM259676v1_genomic.fna.gz && \
    wget -q https://ftp.ncbi.nlm.nih.gov/genomes/all/GCA/002/598/005/GCA_002598005.1_ASM259800v1/GCA_002598005.1_ASM259800v1_genomic.fna.gz

RUN ls /data/test/*.fna.gz > name.txt

#### for highly similar genomes, e.g., > 99.9% ANI, a large sketch size should be used. -s 10204 works well for ANI below 99%.
RUN bindashtree -i name.txt -k 16 -s 10240 -d 1 -t 8 --output_tree try.nwk