houzl3416 / edlmppi Goto Github PK

conda install python==3.8.12
pip install tensorflow==2.4.1
pip install keras==2.4.3
pip install numpy==1.19.5

• Install dependencies

create a program directory
mkdir -p ../programs && cd ../program

• Install SPRINT

git clone https://github.com/lucian-ilie/SPRINT.git
git checkout DELPHI_Server
make compute_HSPs_parallel

• Install psiblast: 2.6.0+ and download the corresponding nr database. (The database used in EDLMPPI is Uniref90)

For Ubuntu:
sudo apt-get install ncbi-blast+

• Intall hh-suite. The database used in DELPHI is uniprot20_2015_06.

• Install GENN+ASAquick

• Install ANCHOR

• Run the following code to extract MBF

  bash feature_computation/compute_features.sh $INPUT_FN
  

• Install virtual environment

  conda create -n ProtT5 pyhton=3.7
  conda activate ProtT5
  

• Following the steps showing in the ProtT5-XL-UniRef50.ipynb

• Notes:

  The result will be saved as .npy

• Install virtual environment

  conda create -n PPI 
  conda activate PPI
  conda install python==3.8.12
  pip install tensorflow==2.4.1
  pip install keras==2.4.3
  pip install numpy==1.19.5
  

• Download the Train Data at http://www.edlmppi.top:5002/download_train

• Following the steps showing in the /train/run.ipynb

• Install virtual environment

  conda create -n PPI 
  conda activate PPI
  conda install python==3.8.12
  pip install tensorflow==2.4.1
  pip install keras==2.4.3
  pip install numpy==1.19.5
  

• The Predict Example can be download in current repository

  python ./predict/pred.py
  

• If you want to get predicted result faster, you can go to http://www.edlmppi.top:5002/ using our online predicted tool.

• If you have any question, you can contact [email protected]

链接: https://pan.baidu.com/s/1UdpI5yENzCxgdd3ZSq66Ng?pwd=52jt 提取码: 52jt 复制这段内容后打开百度网盘手机App，操作更方便哦