Clair3 GPU

clair3-gpu/Dockerfile

@@ -0,0 +1,111 @@
+# syntax=docker/dockerfile:1.4
+# ---------------------------------------------------------------------------
+# Clair3 GPU Dockerfile
+#
+# Requires: NVIDIA Container Toolkit on the host
+# Run with: docker run --gpus all hkubal/clair3:gpu ...
+#
+# CUDA 12.1 requires NVIDIA driver >= 530.30.02
+# To use a different CUDA version, change the base image tag and the
+# --index-url in the PyTorch install step (e.g. cu124 for CUDA 12.4).
+# ---------------------------------------------------------------------------
+FROM nvidia/cuda:11.6.2-runtime-ubuntu20.04
+
+ENV LANG=C.UTF-8 LC_ALL=C.UTF-8
+ENV DEBIAN_FRONTEND=noninteractive
+WORKDIR /opt/bin
+
+# Ensure the NVIDIA runtime exposes GPUs inside the container.
+ENV NVIDIA_VISIBLE_DEVICES=all
+ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
+
+# Install system dependencies and Miniforge (mamba).
+RUN apt-get update && apt-get install -y --no-install-recommends \
+      wget bzip2 ca-certificates && \
+    rm -rf /var/lib/apt/lists/* && \
+    wget -qO /tmp/miniforge.sh \
+      "https://github.com/conda-forge/miniforge/releases/latest/download/Miniforge3-Linux-x86_64.sh" && \
+    bash /tmp/miniforge.sh -b -p /opt/conda && \
+    rm /tmp/miniforge.sh
+
+ENV PATH=/opt/conda/bin:${PATH}
+
+# Step 1 in README: create a conda env with bioinformatics/build dependencies.
+RUN mamba create -n clair3_v2 \
+      -c conda-forge \
+      -c bioconda \
+      python=3.11 \
+      samtools \
+      whatshap \
+      parallel \
+      zstd \
+      xz \
+      zlib \
+      bzip2 \
+      automake \
+      make \
+      gcc \
+      gxx \
+      binutils \
+      curl \
+      pigz \
+      boost-cpp \
+      -y && \
+    mamba clean --all -y
+
+ENV CONDA_PREFIX=/opt/conda/envs/clair3_v2
+ENV CONDA_DEFAULT_ENV=clair3_v2
+ENV PATH=${CONDA_PREFIX}/bin:/opt/bin:/opt/conda/bin:${PATH}
+
+# Step 2 & Step 4 in README: install uv, PyTorch (CUDA 12.1), and Python dependencies.
+RUN pip install --no-cache-dir uv && \
+    uv pip install --python ${CONDA_PREFIX}/bin/python \
+      torch torchvision \
+      --index-url https://download.pytorch.org/whl/cu116 && \
+    uv pip install --python ${CONDA_PREFIX}/bin/python \
+      numpy \
+      h5py \
+      hdf5plugin \
+      numexpr \
+      tqdm \
+      cffi \
+      torchmetrics
+
+# Clair3 source tree
+WORKDIR /
+RUN wget https://github.com/HKU-BAL/Clair3/archive/refs/tags/v2.0.0.tar.gz \
+    && tar -xvzf v2.0.0.tar.gz \
+    && mv Clair3-2.0.0/* /opt/bin
+WORKDIR /opt/bin
+
+# Build Clair3 native components.
+RUN make PREFIX=${CONDA_PREFIX} PYTHON=${CONDA_PREFIX}/bin/python && \
+    cd /opt/bin/preprocess/realign && \
+    g++ -std=c++14 -O1 -shared -fPIC -o realigner ssw_cpp.cpp ssw.c realigner.cpp && \
+    g++ -std=c++11 -shared -fPIC -o debruijn_graph -O3 debruijn_graph.cpp && \
+    rm -rf /opt/bin/samtools-* /opt/bin/longphase-*
+
+# Step 5 in README: install pypy3.11 and mpmath for pypy.
+RUN wget -q https://downloads.python.org/pypy/pypy3.11-v7.3.20-linux64.tar.bz2 && \
+    tar -xjf pypy3.11-v7.3.20-linux64.tar.bz2 && \
+    rm pypy3.11-v7.3.20-linux64.tar.bz2 && \
+    ln -sf /opt/bin/pypy3.11-v7.3.20-linux64/bin/pypy3 ${CONDA_PREFIX}/bin/pypy3 && \
+    ln -sf /opt/bin/pypy3.11-v7.3.20-linux64/bin/pypy3 ${CONDA_PREFIX}/bin/pypy && \
+    pypy3 -m ensurepip && \
+    pypy3 -m pip install --no-cache-dir mpmath==1.2.1
+
+# Step 6 in README: recursively download pre-trained PyTorch models from a directory URL.
+# Pass --build-arg MODEL_CACHE_BUST=$(date +%s) to force a re-download without --no-cache.
+ARG CLAIR3_MODELS_URL=https://www.bio8.cs.hku.hk/clair3/clair3_models_pytorch/
+ARG MODEL_CACHE_BUST=0
+RUN set -eux; \
+    echo "MODEL_CACHE_BUST=${MODEL_CACHE_BUST}"; \
+    mkdir -p /opt/models /tmp/clair3-models; \
+    base_url="${CLAIR3_MODELS_URL%/}"; \
+    wget -r -np -nH --cut-dirs=2 -R "index.html*" -P /tmp/clair3-models "${base_url}/"; \
+    if [ -d /tmp/clair3-models/clair3_models_pytorch ]; then \
+        cp -a /tmp/clair3-models/clair3_models_pytorch/. /opt/models/; \
+    else \
+        cp -a /tmp/clair3-models/. /opt/models/; \
+    fi; \
+    rm -rf /tmp/clair3-models

clair3-gpu/README.md

@@ -0,0 +1,4 @@
+# Clair3-gpu
+
+This image contains the [Clair3](https://github.com/hkubal/Clair3) tool compiled against a specific Cuda runtime version.  
+

clair3-gpu/image.env

@@ -0,0 +1,2 @@
+IMAGE_NAME=clair3-gpu
+IMAGE_TAG=2.0.0-cuda11.6