Rebuilding of Docker image from https://hub.docker.com/r/francecosta/proteinmpnn
Run ProteinMPNN
Run with docker: (Note: Suggested use of --entrypoint bash is not necessary)
docker run --rm -it --platform linux/amd64 jysgro/proteinmpnn:latest
Run with singularity:
singularity pull proteinmpnn.sif docker://jysgro/proteinmpnn
singularity exec ./proteinmpnn.sif python /ProteinMPNN/protein_mpnn_run.py --help
Built on a Macintosh with Silicon Chip M1Pro. Hence to make sure all works added --platform=linux/amd64 in the FROM statement.
Dockerfile:
# SOURCE: https://hub.docker.com/r/francecosta/proteinmpnn
# FROM continuumio/miniconda3:22.11.1
FROM --platform=linux/amd64 continuumio/miniconda3:22.11.1
LABEL version="v0.0.1" maintainer="Francesco Costa <[email protected]>"
RUN git clone https://github.com/dauparas/ProteinMPNN.git
RUN conda create --name mlfold
RUN . ~/.bashrc && \
/opt/conda/bin/conda init bash
RUN echo "source activate mlfold" > ~/.bashrc
ENV PATH /opt/conda/envs/mlfold/bin:$PATH
SHELL ["conda", "run", "-n", "mlfold", "/bin/bash", "-c"]
RUN conda install pytorch torchvision torchaudio cudatoolkit=11.3 -c pytorch
ENV PATH /mlfold/:$PATH
WORKDIR /ProteinMPNN
Content type
Image
Digest
sha256:0d4a89762…
Size
2.7 GB
Last updated
over 2 years ago
docker pull jysgro/proteinmpnn