Large-scale predicting protein functions through multi-level feature fusion

PredGO, a protein gene ontology prediction method that fuses multi-level features including sequence, predicted structure and protein-protein interaction information.

Usage Notes: You can download the Alphafold predicted PDB file from AlphaFoldDB or UniProtKB as input. The upload PDB file should be a monomer (Only the first chain in the PDB file will be used).

                        The system will call the MFO, BPO, CCO models trained on the UNIGOA16 dataset to predict GO terms and then combine their results.

  • sequence search

Enter Query Sequence

upload a FASTA file

Or, input FASTA sequence(s)(for example: A0A2K4XGN6, Q47319), maximum allowed 5 sequences at one time

Email Address:      Optional

Submission Title:   Optional

Private Key:          Secure code for private access only

 

 

Please click this link to download the data.

For questions about this website, please email for assistance.

Last modified 7/1/2023.