安裝之前注意gcc版本，我用centos8的4.8.5裝不了，后面自己安裝的gcc8.4.0
1.
conda create -n af2 python=3.8 -y
2.
conda activate af2
3.
conda install -y -c nvidia cudnn==8.0.4
4.
conda install -y -c bioconda hmmer==3.3.2 kalign2==2.04
5.
conda install -c conda-forge -c bioconda hhsuite==3.3.0
6.
conda install -y -c conda-forge openmm=7.5.1 pdbfixer==1.7 pip
7.
pip3 install --upgrade pip && pip3 install -r ./requirements.txt
pip3 install "jax[cuda111]" -f https://storage.googleapis.com/jax-releases/jax_cuda_releases.html
注：如果這步的jaxlib-0.1.69+cuda111沒有裝成功會報找不到gpu的錯誤（conda list | grep jax）。
8.
work_path=/home/jianghf/Programs/alphafold2
9.
cd ~/Programs/anaconda3/envs/af2/lib/python3.8/site-packages/
10.
patch -p0 < $work_path/docker/openmm.patch
運行命令：
run_alphafold.py --fasta_paths CYP71A1.fasta --output_dir .