DeepComplex2

Distance-based methods to generate complex structures using inter-chain residue-residue contacts. This package includes two optimization tools to reconstruct the quaternary structures of protein dimers: a gradient descent (GD) method and a Markov Chain Monte Carlo (MC) method. The two methods depend on PyRosett4 package.

(1) Download the code (short path is recommended)

git clone [email protected]:jianlin-cheng/DeepComplex2.git

(If fail, try username) git clone https://github.com/jianlin-cheng/DeepComplex2.git

(2) Install PyRosetta4: http://www.pyrosetta.org/dow

(3) Run the code for generating protein complexes using gradient descent method

   Usage:
   $ sh scripts/run_dock_gd.sh <target id> <initial pdb file>.pdb <path of restraint file> <output folder> <path of weight file>

   Example:
   $ sh scripts/run_dock_gd.sh 1D3Y  /data/esdft/initial_starts/1D3Y_modified.pdb  /data/esdft/restraints/1D3Y_AB.rr  /data/esdft/output  /data/esdft/weight_files/talaris2013.wts

(4) Run the code for generating protein complexes using Markov chain Monte Carlo

   Usage:
   $ sh scripts/run_dock_mc.sh <target id> <initial pdb file>.pdb <path of restraint file> <output folder> <path of weight file>

   Example:
   $ sh scripts/run_dock_mc.sh 1D3Y  /data/esdft/initial_starts/1D3Y_modified.pdb  /data/esdft/restraints/1D3Y_AB.rr  /data/esdft/output  /data/esdft/weight_files/talaris2013.wts

A related tool of reconstructing complexe structures using CNS from inter-chain contacts

To Generate complex structures by CNS run the scripts at: https://github.com/jianlin-cheng/DeepComplex.