From 4eeb635070c9e67323ad531cddc118ce02df49d0 Mon Sep 17 00:00:00 2001
From: Alex Petty <pettyalex@gmail.com>
Date: Wed, 21 Aug 2024 12:56:58 -0400
Subject: [PATCH] Link htslib to libdeflate, improving performance. (#1022)

---
 htslib/1.20.c/Dockerfile | 97 ++++++++++++++++++++++++++++++++++++++++
 htslib/1.20.c/README.md  | 26 +++++++++++
 2 files changed, 123 insertions(+)
 create mode 100644 htslib/1.20.c/Dockerfile
 create mode 100644 htslib/1.20.c/README.md

diff --git a/htslib/1.20.c/Dockerfile b/htslib/1.20.c/Dockerfile
new file mode 100644
index 000000000..bc1b1d8ed
--- /dev/null
+++ b/htslib/1.20.c/Dockerfile
@@ -0,0 +1,97 @@
+# for easy upgrade later. ARG variables only persist during build time
+ARG HTSLIB_VER="1.20"
+
+FROM ubuntu:jammy as builder
+
+ARG HTSLIB_VER
+
+# install dependencies, cleanup apt garbage
+# It's helpful when they're all listed on https://github.com/samtools/htslib/blob/develop/INSTALL
+RUN apt-get update && apt-get install --no-install-recommends -y \
+    wget \
+    ca-certificates \
+    make \
+    bzip2 \
+    autoconf \
+    automake \
+    make \
+    gcc \
+    perl \
+    zlib1g-dev \
+    libbz2-dev \
+    liblzma-dev \
+    libcurl4-gnutls-dev \
+    libssl-dev \
+    libdeflate-dev \
+    procps && \
+    rm -rf /var/lib/apt/lists/* && apt-get autoclean
+
+# get htslib, compile, install, run test suite
+RUN wget -q https://github.com/samtools/htslib/releases/download/${HTSLIB_VER}/htslib-${HTSLIB_VER}.tar.bz2 && \
+    tar -vxjf htslib-${HTSLIB_VER}.tar.bz2 && \
+    rm -v htslib-${HTSLIB_VER}.tar.bz2 && \
+    cd htslib-${HTSLIB_VER} && \
+    ./configure && \
+    make && \
+    make install && \
+    make test
+
+### start of app stage ###
+FROM ubuntu:jammy as app 
+
+ARG HTSLIB_VER
+
+LABEL base.image="ubuntu:jammy"
+LABEL dockerfile.version="1"
+LABEL software="htslib"
+LABEL software.version="${HTSLIB_VER}"
+LABEL description="A C library for reading/writing high-throughput sequencing data"
+LABEL website="https://github.com/samtools/htslib"
+LABEL license="https://github.com/samtools/htslib/blob/develop/LICENSE"
+LABEL maintainer="Erin Young"
+LABEL maintainer.email="eriny@utah.gov"
+LABEL maintainer2="Curtis Kapsak"
+LABEL maintainer2.email="kapsakcj@gmail.com"
+
+# install runtime dependencies & cleanup apt garbage
+# installed as recommend here: https://github.com/samtools/htslib/blob/develop/INSTALL#L31
+RUN apt-get update && apt-get install --no-install-recommends -y \
+    bzip2 \
+    zlib1g \
+    libbz2-1.0 \
+    liblzma5 \
+    libcurl3-gnutls \
+    libdeflate0 \
+    ca-certificates \ 
+    && apt-get autoclean && rm -rf /var/lib/apt/lists/*
+
+# copy in htslib executables from builder stage
+COPY --from=builder /usr/local/bin/* /usr/local/bin/
+COPY --from=builder /usr/local/lib/ /usr/local/lib/
+COPY --from=builder /usr/local/include/ /usr/local/include/
+
+# set locale settings for singularity compatibility
+ENV LC_ALL=C
+
+# set working directory
+WORKDIR /data
+
+# default command is to show help options
+CMD ["htsfile", "--help"]
+
+### start of test stage ###
+FROM app as test
+
+# check that these three executables are available
+RUN bgzip --help && tabix --help && htsfile --help
+
+RUN apt-get update && apt-get install --no-install-recommends -y wget
+
+# use on actual files
+RUN wget -q https://github.com/StaPH-B/docker-builds/raw/master/tests/SARS-CoV-2/SRR13957123_1.fastq.gz && \
+    gunzip SRR13957123_1.fastq.gz && \
+    bgzip SRR13957123_1.fastq
+
+# FYI Test suite "make test" now performed in the builder stage since app and
+# test stages do not include htslib source code.
+# This is to avoid having to re-download source code simply to run test suite
\ No newline at end of file
diff --git a/htslib/1.20.c/README.md b/htslib/1.20.c/README.md
new file mode 100644
index 000000000..7d03b6333
--- /dev/null
+++ b/htslib/1.20.c/README.md
@@ -0,0 +1,26 @@
+# htslib container
+
+Main tool:
+
+* [https://www.htslib.org/](https://www.htslib.org/)
+* [GitHub](https://github.com/samtools/htslib)
+
+Additional tools:
+
+* perl 5.34.0
+
+## Example Usage
+
+```bash
+# determine file formats for various BAM and SAM files
+$ htsfile tests/SARS-CoV-2/SRR13957123.primertrim.sorted.bam 
+tests/SARS-CoV-2/SRR13957123.primertrim.sorted.bam:     BAM version 1 compressed sequence data
+
+$ htsfile ce_tag_padded.sam
+ce_tag_padded.sam:      SAM version 1.4 sequence text
+
+# compresses sample.fastq to sample.fastq.gz in BGZF format (blocked GNU Zip Format)
+$ bgzip sample.fastq
+```
+
+Better documentation can be found at [https://www.htslib.org/doc/samtools.html](https://www.htslib.org/doc/samtools.html)