PANDA2: Protein function prediction using graph neural network

This repository includes code and a pre-trained model of PANDA2 for protein function prediction.

Stand-alone PANDA2

1. Download the trained model

wget http://dna.cs.miami.edu/PANDA2/download_files/gcn37onlyesm_cafa.cnn ./
wget http://dna.cs.miami.edu/PANDA2/download_files/gcn37onlyesm_cafa.gcn ./
wget https://dna.cs.miami.edu/PANDA2/download_files/data-cafa.zip ./
unzip data-cafa.zip

3. Run PANDA2 with a fasta format file

# perl parse_seq_cut.pl $input_fasta $output_dir
>> perl parse_seq_cut.pl example/test_5_samples.fasta example/test_5_samples.
# python panda2_features.py $output_dir
# a $output_dir.2 will automaticly created.
>> python panda2_features.py example/test_5_samples
# make prediction with the feature file
>> python panda2_prediction.py example/test_5_samples.2/panda2_features.pkl

Output explaination:

The output is saved in example/test_5_samples.2/panda2_prediction.txt. The file format is as follows:

• The first two lines contain model information.
• Subsequent lines follow this format:
  • The first column is the name of the PDB structure file.
  • The second column is the GO term ID.
  • The third column is the confidence score predicted by PANDA2.
• The last line contains only "END."

AUTHOR PANDA2
MODEL 1
KEYWORDS graph network, sequence alignment.
T100900000046   GO:0097458      0.49
T100900000046   GO:0003674      0.91
...
T100900000026   GO:0006887      0.16
T100900000026   GO:0046578      0.09
END

Dependencies
The conda environment is shared via "environment.yml".
In order to locally run PANDA2, you also need to install blast-2.2.23 and then update $blast_path in panda2_psiblast.py.

Web-server PANDA2

Submit jobs at http://dna.cs.miami.edu/PANDA2/

Citation

@article{10.1093/nargab/lqac004,
    author = {Zhao, Chenguang and Liu, Tong and Wang, Zheng},
    title = "{PANDA2: protein function prediction using graph neural networks}",
    journal = {NAR Genomics and Bioinformatics},
    volume = {4},
    number = {1},
    pages = {lqac004},
    year = {2022},
    month = {02},
    issn = {2631-9268},
    doi = {10.1093/nargab/lqac004},
}