RFdiffusion学习笔记

RFdiffusion介绍

TODO

RFdiffusion安装

1. 安装准备工作

git下载：

git clone https://github.com/RosettaCommons/RFdiffusion.git

进入目录并下载模型的weights：

cd RFdiffusion
mkdir models && cd models
wget http://files.ipd.uw.edu/pub/RFdiffusion/6f5902ac237024bdd0c176cb93063dc4/Base_ckpt.pt
wget http://files.ipd.uw.edu/pub/RFdiffusion/e29311f6f1bf1af907f9ef9f44b8328b/Complex_base_ckpt.pt
wget http://files.ipd.uw.edu/pub/RFdiffusion/60f09a193fb5e5ccdc4980417708dbab/Complex_Fold_base_ckpt.pt
wget http://files.ipd.uw.edu/pub/RFdiffusion/74f51cfb8b440f50d70878e05361d8f0/InpaintSeq_ckpt.pt
wget http://files.ipd.uw.edu/pub/RFdiffusion/76d00716416567174cdb7ca96e208296/InpaintSeq_Fold_ckpt.pt
wget http://files.ipd.uw.edu/pub/RFdiffusion/5532d2e1f3a4738decd58b19d633b3c3/ActiveSite_ckpt.pt
wget http://files.ipd.uw.edu/pub/RFdiffusion/12fc204edeae5b57713c5ad7dcb97d39/Base_epoch8_ckpt.pt

Optional:
wget http://files.ipd.uw.edu/pub/RFdiffusion/f572d396fae9206628714fb2ce00f72e/Complex_beta_ckpt.pt

# original structure prediction weights
wget http://files.ipd.uw.edu/pub/RFdiffusion/1befcb9b28e2f778f53d47f18b7597fa/RF_structure_prediction_weights.pt

2. 通过docker执行

由于我安装的CUDA版本为12.1,只能采用docker版本，具体方法如下：
构建镜像

docker build -f docker/Dockerfile -t rfdiffusion .

由于numpy已经更新，需要对numpy进行降级，可以使用如下:

# Usage: 
# git clone https://github.com/RosettaCommons/RFdiffusion.git
# cd RFdiffusion
# docker build -f docker/Dockerfile -t rfdiffusion .
# mkdir $HOME/inputs $HOME/outputs $HOME/models
# bash scripts/download_models.sh $HOME/models
# wget -P $HOME/inputs https://files.rcsb.org/view/5TPN.pdb

# docker run -it --rm --gpus all \
#   -v $HOME/models:$HOME/models \
#   -v $HOME/inputs:$HOME/inputs \
#   -v $HOME/outputs:$HOME/outputs \
#   rfdiffusion \
#   inference.output_prefix=$HOME/outputs/motifscaffolding \
#   inference.model_directory_path=$HOME/models \
#   inference.input_pdb=$HOME/inputs/5TPN.pdb \
#   inference.num_designs=3 \
#   'contigmap.contigs=[10-40/A163-181/10-40]'

FROM nvcr.io/nvidia/cuda:11.6.2-cudnn8-runtime-ubuntu20.04

COPY . /app/RFdiffusion/

RUN apt-get -q update \ 
  && DEBIAN_FRONTEND=noninteractive apt-get install --no-install-recommends -y \
  git \
  python3.9 \
  python3-pip \
  && python3.9 -m pip install -q -U --no-cache-dir pip \
  && rm -rf /var/lib/apt/lists/* \
  && apt-get autoremove -y \
  && apt-get clean \
  && pip install -q --no-cache-dir \
  dgl==1.0.2+cu116 -f https://data.dgl.ai/wheels/cu116/repo.html \
  torch==1.12.1+cu116 --extra-index-url https://download.pytorch.org/whl/cu116 \
  numpy==1.26.4 \
  e3nn==0.3.3 \
  wandb==0.12.0 \
  pynvml==11.0.0 \
  git+https://github.com/NVIDIA/dllogger#egg=dllogger \
  decorator==5.1.0 \
  hydra-core==1.3.2 \
  pyrsistent==0.19.3 \
  /app/RFdiffusion/env/SE3Transformer \
  && pip install --no-cache-dir /app/RFdiffusion --no-deps
  
WORKDIR /app/RFdiffusion

ENV DGLBACKEND="pytorch"

ENTRYPOINT ["python3.9", "scripts/run_inference.py"]

RFdiffusion使用

因为我安装的是docker版本，这里简单介绍docker版本的使用，如下:

docker run -it --rm --gpus all \
  -v $HOME/models:$HOME/models \
  -v $HOME/inputs:$HOME/inputs \
  -v $HOME/outputs:$HOME/outputs \
  rfdiffusion \
  inference.output_prefix=$HOME/outputs/motifscaffolding \
  inference.model_directory_path=$HOME/models \
  inference.input_pdb=$HOME/inputs/5TPN.pdb \
  inference.num_designs=3 \
  'contigmap.contigs=[10-40/A163-181/10-40]'

使用过程可能会报如下错误:

could not select device driver “” with capabilities: [[gpu]].

主要是没有安装nvida-container，官方详细说明可以看此连接
简单安装如下：

1. 加入包索引

curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \
  && curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
    sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
    sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list

2. 安装

   sudo apt-get update
   sudo apt-get install -y nvidia-container-toolkit