1. 下载RoseTTAFold
$ git clone https://github.com/RosettaCommons/RoseTTAFold.git (100M)
$ cd RoseTTAFold
2. 下载权重、数据库
$ wget https://files.ipd.uw.edu/pub/RoseTTAFold/weights.tar.gz (0.9G)
$ tar xfz weights.tar.gz (1.1G)
$ wget http://wwwuser.gwdg.de/~compbiol/uniclust/2020_06/UniRef30_2020_06_hhsuite.tar.gz (47G)
$ mkdir -p UniRef30_2020_06
$ tar xfz UniRef30_2020_06_hhsuite.tar.gz -C ./UniRef30_2020_06 (181G)
$ wget https://bfd.mmseqs.com/bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt.tar.gz (272G)
$ mkdir -p bfd
$ tar xfz bfd_metaclust_clu_complete_id30_c90_final_seq.sorted_opt.tar.gz -C ./bfd (1.8T)
$ wget https://files.ipd.uw.edu/pub/RoseTTAFold/pdb100_2021Mar03.tar.gz (115G)
$ tar xfz pdb100_2021Mar03.tar.gz (667G)
# for CASP14 benchmarks, we used this one: https://files.ipd.uw.edu/pub/RoseTTAFold/pdb100_2020Mar11.tar.gz
3. 安装Conda环境
# create conda environment for RoseTTAFold
$ conda env create -f RoseTTAFold-linux.yml (cuda11) [RoseTTAFold-linux-cu101.yml (cuda10.1)]
$ conda env create -f folding-linux.yml (pyrosetta)
$ ./install_dependencies.sh
$ conda activate folding
$ conda config --add channels https://levinthal:paradox@conda.graylab.jhu.edu
$ conda install pyrosetta=2021.27+release.7ce6488
4. 环境配置
(1). 数据库调用
由于数据库太大（~1.5T），使用公用数据库软链接调用
IP: 172.16.2.207
$ ln -s /data0/wangq/Databases/bfd [target dir]
$ ln -s /data0/wangq/Databases/pdb100 [target dir]
$ ln -s /data0/wangq/Databases/UniRef30 [target dir]
(2). Psipred运行报错
https://www.cnblogs.com/wq242424/p/15037069.html
5. 运行
1 run_[pyrosetta, e2e]_ver.sh input.fa [dir]
LOG:
Running Hhblits
Running PSIPRED
Running hhsearch
Predicting distance and orientations
Running parallel RosettaTR.py
Running DeepAccNet-msa
Picking final models
Final models saved in: ./model
Done