TALOS

Introduction

TALOS (Torsion Angle Likelihood Obtained from Shift and sequence similarity) is a database system for empirical prediction of phi and psi backbone torsion angles from five kinds (HA, CA, CB, CO, N) of chemical shifts for a given protein sequence.

For detailed information please check the TALOS website.

Using TALOS

1. Create a subdirectory (for example, structure/cyana21/talos) and copy the latest sequence and atom list files there. It is convenient to have them named XXXX.seq and XXXX.prot, where XXXX is an NESG target ID or other protein name. When using CARA, export the chemical shifts as an atom list file in this directory.
2. Create and init.cya in this directory as described in "Creating an init.cya file for CYANA 2.1" or copy a previously used file.
3. Start CYANA and type:

     read prot XXXX.prot
     taloslist XXXX

1. This will create the TALOS input file XXXX.tab. In a UNIX shell run talos.tcl -in XXXX.tab. During the database search, a series of files pred/res*.tab will be created. Each one of these files tallies the 10 best database matches for a given residue in the target protein. Before exiting, talos.tcl calls vina.tcl to create a file pred.tab, which includes an initial summary of the prediction results. The database search will typically take about 10-60 sec per residue in the target.
2. In a UNIX shell run rama.tcl -in XXXX.tab. Here you can examine phi and psi distributions, choose database matches to be used in calculating predictions, and classify prediction results as Good, Ambiguous or Unclassified / New. See below for the guidelines for classifying prediction. Save your modifications in a new file, for example, talos.tab.
3. Start CYANA and type:

   talosaco pred    #or "talos.tab" -- use the appropriate filename
   write aco talos.aco

See also the ~/demo/details/TalosAngleRestraints.cya example script in your local CYANA 2.1 installation.

Manual Refinement of TALOS predictions

Guidelines for refining the TALOS output:

• Classify prediction as Good only if
  • All 10 best database matches fall in a "consistent" region of the Ramachandran map
  • Or 9 out of 10 best database matches fall in a consistent region with phi < 0, and the one outlier also lies in phi < 0 half of the map
  • Or 9 out of 10 of the best database matches fall in a consistent region with phi > 0
• Accept predictions which are classified as Good, whose residues are in beta-sheets or helices according to CSI (excluding the first and the last residue of a secondary structure element).

For de novo structure determination it is recommended to take the automatically generated TALOS constraints. Angular constraints outside of secondary structure elements (as determined by CSI) can be commented out in the talos.aco file.

During structure refinement you can refine TALOS predictions against a preliminary structure.

vina.tcl -in XXXX.tab -ref XXXX.pdb -auto

and

rama.tcl -in XXXX.tab -ref XXXX.pdb

The XXXX.pdb file must have only one conformer. Thus, you may need to analyze the angle distributions in a molecular graphics package (e.g. MOLMOL).

Element	PHI	PSI
a-helix	-60	-45
b-sheet	-140	135

Description of talosaco.cya Macro

Invoked as:

talosaco file [factor [width]]

Here file is the TALOS prediction output, width is the threshold minimum width for PHI/PSI angle distributions, and factor is used to scale the width of a distribution when creating an angle constraint. Both width and factor arguments are optional. By default, width=20.0 and factor=2.0.

This macro will create angle constraints for a given residue only if the prediction is classified as "Good" and the residue is not a proline.

References

-- AlexEletski - 7 Jun 2007

-- Updated by JimAramini - 11 Nov 2009