Modeling disordered protein interactions from biophysical principles
Clone or download
lennax and GitHub Enterprise Update
Latest commit 0ffdf79 Jul 19, 2017
Type Name Latest commit message Commit time
Failed to load latest commit information.


IDP-LZerD models the bound conformation of a disordered PPI, where one intrinsically disordered protein (IDP) binds to an ordered protein. The inputs are the sequence of a disordered protein and the structure of an ordered protein.

Copyright (C) 2016-2017 Lenna X. Peterson, Daisuke Kihara, and Purdue University.

License: GPL v3 for academic use. (For commercial use, please contact us for different licensing)

Contact: Daisuke Kihara (

Reference: Lenna Peterson, A Roy, C Christoffer, G Terashi, and D Kihara. (2017) Modeling disordered protein interactions from biophysical principles. PLoS Computational Biology 13: e1005485. doi: 10.1371/journal.pcbi.1005485


To run the test, LZerD and all Python dependencies are required.

Python dependencies

  • apsw
  • numpy
  • scipy
  • pandas
  • Biopython

Binary dependencies

Getting started

  1. Edit PATHS.ini to specify path to LZerD, blastpgp, nr, and Rosetta

Test protein

  1. Edit 'test/' to specify path to IDP-LZerD
  2. Run test (NOTE: creates ~250 GB of files): cd test && ./
  3. Non-refined paths will be output in 'test/4ah2ac3'
  4. Follow standard procedures to prepare PDB files for CHARMM
  5. Edit CHARMM input file 'idp_relax.inp' to run CHARMM

Running a new protein

  1. Prepare FASTA sequence of IDP and structure of ordered partner protein
  2. Create fragments with Rosetta using rosetta_tools/fragment_tools/ and the flags '-frag_sizes 9 -n_frags 30'
  3. Convert Rosetta output to PDB format using 'scripts/'
  4. Rename 'window_data.csv' to '${PDBID}_data.csv'
  5. Convert CA to full-atom backbones using Pulchra
  6. Build side-chains using chosen side-chain modeling software
  7. Put fragments into subdirectories as in 'test/4ah2'
  8. Run LZerD on each fragment
  9. Score LZerD output using GOAP and ITScorePro
  10. Copy and edit 'test/' for new protein and run as for test protein