Skip to content
Switch branches/tags

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


Distance- and orientation-based Covariational threadER

Download and Installation

$ git clone
$ cd DisCovER
$ g++ -O3 -std=c++0x -o DisCovER DisCovER.cpp 

The DisCovER executable (named as DisCovER) is compiled and tested on x86_64 Linux system.


To see the usage instructions, run $ ./DisCovER -h

!!! HELP !!!

 DisCovER : Distance- and orientation-based Covariational threadER (V1.0)
 For comments, please email to

Usage:   ./DisCovER [-option] [argument]

Options:  -T path to template library             	- input, required
          -L .txt file with template IDs 	 	- input, required
          -q query ID   		          	- input, required
          -o path to output directory              	- input, required
          -n number of TOP templates to be selected  	(optional, default 50)
          -c sequence identity cutoff	             	(optional, default 0.30)
More details on some options:
  • -T Path to template directory, containing PDB, MTX, DEP, dssp, and seq sub-directories. PDB: Non-redundant PDB structures, MTX: Sequence Profiles from PSI-BLAST search, DEP: Depth-dependent structure profiles, dssp: DSSP generated output files, seq: FASTA sequence (Ref: Please make sure your template directory path ends with "/" (e.g. /home/template_library/) to avoid Segmentation Error.
  • -L Path to the file containing template (pdb) IDs (sample template_list.txt file in DisCovER_template_Lib.tar.gz provided for your reference.)

Test DisCovER

To run DisCovER, go to the target directory, which contains target sequence file, standard SPIDER3 predicted output file, profiles, and inter-residue distance and orientations. Particularly, DisCovER will assume seq.fasta, seq.spd33, seq.prf, seq.mtx, query.rr, omega.txt, theta.txt, phi.txt files are prersent in the current directory.

$ ulimit -s 419430400                        (To increase the stack limit, avoiding Segmentation Error)
$ cd Example
$ tar -zxvf DisCovER_template_Lib.tar.gz           (Uncompress a toy template library for the test run)
$ cd 6GW7_A/                     (Target directory contains respective input files as mentioned above)
# run DisCovER 
$ ../../DisCovER -T ../template_library/ -L ../template_library/template_list.txt -q 6GW7_A -o ./ -n 50 -c 1.0  

The predicted top-one alignment file is named as top1.fasta. The alignmnet file for 6GW7_A is given here for your reference. When you run DisCovER, the output screen looks like it.

Generating input files from Query sequence

  • Generate an alignmnet file (.aln format) of the query sequence.
  • Get trRosetta output file (.npz).
  • Go through the steps discussed here.

Generating template library

Building 3D models

  • The standard automodel() class of MODELLER is available here.
  • The MODELLER with additional restraints from predicted distances, orientations, and secondary structures script is available here. Go through the usage instruction here. The helping scripts to convert trRosetta output file (.npz) to inter-residue geometries are available here.


  • The benchmark datasets along with input files can be found here
  • DisCovER predicted query-template alignment files can be found here
  • DisCovER predicted full-length 3D models can be found here


If you find DisCovER useful, please cite our PROTEINS paper.


No releases published


No packages published