inter-protein contact prediction from sequence.

• python3.8

  1. pytorch1.9
  2. Biopython
  3. esm
  4. numpy

• other packages

  1. alnstats
  2. fasta2aln
  3. hh-suite
  4. CCMpred

git clone https://github.com/ChengfeiYan/DRN-1D2D_Inter.git

Download the trained models from trained models, then unzip it into the folder named "model".

python predict.py sequenceA msaA sequenceB msaB result_path device

Where MSA should be derived from Uniref90 or Uniref100 database.

python predict.py ./example/1GL1_A.fasta ./example/1GL1_A_uniref100.a3m ./example/1GL1_I.fasta ./example/1GL1_I_uniref100.a3m ./example/result cpu

Please cite: Improved inter-protein contact prediction using dimensional hybrid residual networks and protein language models Yunda Si, Chengfei Yan bioRxiv 2022.08.04.502748; doi: https://doi.org/10.1101/2022.08.04.502748.

If you meet any problem in installing or running the program, please contact [email protected].