visualizing docking output
- From: Pat Walters <pwalters()at()vpharm.com>
- Organization: Vertex Pharmaceuticals
- Subject: visualizing docking output
- Date: Wed, 14 Jul 1999 09:17:25 -0400
There have a been a few recent posts regarding visualization of the
output from docking calculations so thought I'd throw in my 0.02.
While it is relatively easy to visualize a single protein ligand complex
any of a number of tools, most of us are not interested in looking at
just one ligand. Typically the user of a docking program will want to
examine dozens or even hundreds of docked ligands. To my knowledge
there is currently no commercial or academic software which adequately
addresses this problem. Before software vendors get upset and start
flaming me, let me specify what I think is necessary. The program
should be able to
1. Read a protein as a PDB file.
2. Read a multi-structure file containing a set of ligands in a file
format appropriate for small molecules (SDF,mol2, not PDB).
3. Allow facile creation and display of a protein surface colored by
potential. It should also be easy to toggle the surface on and off.
4. Be capable of displaying hydrogen bonds between the protein and
5. Allow the user to easily scroll through the set of ligands,
each ligand in the active site of the protein.
6. Provide adequate performance with macromolecular systems.
It would also be nice to be able to
7. Mark interesting compounds and save the marked compounds to a
8. Display a 2D ligand structure along with the 3D structure.
9. Display docking scores along with 3D strucutres.
10. Filter structures - e.g. show me only those structure which make a
hydrogen bond with ASP 87.
11. Visualize close contacts.
None of this is rocket science, but I have yet to find a package which
satisfies all the requirements. Our solution was to write our own
viewer. Unfortunately this viewer relies on Molecular Inventor which has
been orphaned by SGI
I'd be happy if anyone could prove me wrong and point me to software
which fulfills the requirements above. I'd also be interested in any
Active-X/COM/WhateverMicrosoftIsCallingItToday components for
doing molecular graphics. I'll gladly summarize for the list.
W. Patrick Walters, Ph.D.
Vertex Pharmaceuticals, Inc., 130 Waverley St., Cambridge, MA 02139
Voice: (617)577-6242 FAX: (617)577-6400
friends don't let friends write FORTRAN code.