Rosetta 3.1 Release Manual

Cluster Application

This is a Clustering tool for PDBs and or silent files in structure predictions. This tool helps to identify the decocy with the most neighbors within a certain RMSD threshold by clustering bunch of protein structures.The cluster application is at rosetta_source/src/apps/public/. The name of the application is cluster.cc

Command Lines and Options

• Sample command:

  cluster -database ~/rosetta_database -in:file:silent silent.out -in::file::nonideal_silentfile -in::file::fullatom
  -native 1a19.pdb > cluster.log
  

• Options:

  * PDB input:
  ------------
  -in:file:s *.pdb   or
  -in:file:l  list_of_pdbs
  -no_optH                                    Dont change positions of Hydrogen atoms. [Boolean]
  
  *Silent input:
  --------------
  -in:file:silent                             Silent input filesname.[String]
  -in:file:s                                  specify specific tags to be extracted, if left out all will be taken
  -in:file:fullatom                           For full atom structures. [Boolean]
  
  *Native:
  --------
  -in:file:native                             Native PDB if CaRMS is required
  
  *Scorefunction:
  -score:weights  			                      Name of weight set or weights file, default is 'standard'. [String]
  -score:patch                                Name of patch file (without extension)" , default="". [String]
  -rescore:verbose                            Full break down of weights, raw scores and weighted scores. [Boolean]
  -rescore:output_only
  
  *Output:
  -------
  -nooutput                                   Don't print PDB structures (default false) [Boolean]
  -out:prefix                                 Prefix the output structures with a string.[String]
  
  *Clustering:
  ------------
  -input_score_filter                         Ignore structures above certain energy
  -exclude_res                                Exclude residue numbers
  -radius                                     Cluster radius
  -limit_cluster_size                         Maximal cluster size
  -limit_clusters                             Maximal number of clusters
  -limit_total_structures                     Maximal number of structures in total
  -limit_dist_matrix                          <defunc> only pre cluster N structures, then redo.
  -sort_groups_by_energy                      Sort clusters by energy.
  -remove_highest_energy_member               Remove highest energy member of each cluster
  -cluster:radius                             RMSD in Angstrom
  

Result Interpretation

You can use -nooutput to skip generating pdb files. All the clustering information are recored in the log file. You can find the "Summary" section at the bottom of the log file. For Example:

---------- Summary ---------------------------------
Cluster:  0  N: 1   c.0.*.pdb
    1HOE_0001  1.44308 0

Cluster:  1  N: 1   c.1.*.pdb
    1QYS_0001  97.4857 0

----------------------------------------------------
  Clusters: 2
  Structures: 2
----------------------------------------------------
Timing:
  Readin:3s
  Cluster: 0s
  Additional Clustering: 0s
  Total: 3

---

© Copyright Rosetta Commons Member Institutions. For more information, see http://www.rosettacommons.org.