Evolution of human-virus PPI interfaces

Building a human-virus structural interaction network (SIN), resolving protein-protein interaction (PPI) interfaces, mapping and categorizing interfaces, and computing the evolutionary rates of different PPI interfaces.

Associated paper (by Su, W.-C. & Xia, Y.): Virus targeting as a dominant driver of interfacial evolution in the structurally resolved human-virus protein-protein interaction network.

Getting started

1. Clone this repository.
   $ git clone 'https://github.com/wanchunsu/HV_PPI_interface_evolution.git'

2. Navigate to the downloaded folder.
   $ cd HV_PPI_interface_evolution

3. Install requirements listed in dependencies.txt.

Usage

1. Navigate to the scripts directory.
   $ cd scripts

2. Follow the instructions in the instructions.bash script. Below are the main steps in this pipeline:

   1. Build a human-virus SIN and resolve interaction interfaces.
   2. Map interfacial residues back onto human target proteins and categorize them.
   3. Determine the evolutionary rates of different PPI interfaces.

Note: The working directory is scripts and all data and results are stored in the data directory.

Software

• Python 3.9.12
• BLAST 2.13.0 (see https://www.ncbi.nlm.nih.gov/books/NBK52640/ for installation guide)
• DSSP (mkdssp) (installed in the install_dssp.bash script)
• MAFFT v7.490 (installed in the construct_msas.bash script)
• PAML (installed in the get_binned_dnds.bash script)
• PAL2NAL v14 (installed in the calc_site_specific_dnds.bash script)
• HyPhy 2.5.40 (installed in the calc_site_specific_dnds.bash script)

Databases

• IntAct (https://www.ebi.ac.uk/intact/home)
• PDB (https://www.rcsb.org/)
• UniProt (https://www.uniprot.org/)
• Ensembl (https://useast.ensembl.org/index.html)
• RefSeq (https://www.ncbi.nlm.nih.gov/refseq/)
• TimeTree (http://timetree.org/)

Contact

Wan-Chun Su wan.su@mail.mcgill.ca