Questions about the molecular docking problem

S. Park separk at pollux.usc.edu
Tue Nov 28 07:41:10 EST 1995


Hello,

I have been studying the rigid molecular docking problem and have several
questions about the assumptions typically made.  Since my major is computer
science, I lack lots of basic knowledge on the subject and I wonder if 
anyone could kindly tell me about your opinion and give me any references
that I could use for the study.  I would really appreciate it !

Here are my questions:

It seems to me that there are three assumptions for the molecular docking 
problem:

1. Rigidity
2. Availability of the atomic coordinates
3. Solvent effects related assumption

First of all, I would like to know how "accurate" the results we have with
the assumptions above.  For example, I understand that the answers are
"approximate" solutions for the real answer.  But how reliable (or,
exact, valid ?) are they ?  Can we quantify this ?

Secondly, it seems to me that most of the automated docking procedures
make use of the atomic coordinates of the molecules.  But what if we don't
have their crystal structures ?  How could we find the mode of binding in
these cases ?

Thirdly, I wonder how much information could we glean from the geometric
complementarity based approach.  I understand that this is a sort of
"screening" process for later "energy minimization".  It seems to me that
geometric complementarity serves as a necessary condition for the real
answer.  But what if we miss the real "answer" in this "screening" step ?

Finally, I wonder how reliable the result that comes from without considering
the surrounding solvent is.  It seems to me that one of the reasons that we
don't consider the solvent effect is due to the expensive computational cost.
But then how "accurate" is the result ?

Thank you very much for your time !

Regards,
Seongbin Park (separk at pollux.usc.edu)




More information about the Molmodel mailing list