On Mon, Apr 12, 2021, Manisha wrote:
>I did run the simulations for quite long almost 500ns. But the main
>question here is Why the stretch of DNA which shows binding in Anisotropy
>assay is unable to show binding in MMPBSA and also the stretch which shows
>absolutely no binding in anisotropy assay shows significant binding in
>MMPBSA?
The are many reasons why a simulation may disagree with experiment:
* Your starting structures may not adequately represent how the ligand binds
* There may be insufficient sampling
* Force fields may not correctly represent the real interactions
* The implicit solvent model in MMPBSA may not be accurate enough
* The solvent/environment conditions in the simulation may not represent
those in the experiment
* You may have made an error in carrying out the simulation
* You (or the original authors) may be interpreting the experimental results
in an incorrect fashion
* etc.
Remember the goal of a simulation is (generally) not just to reproduce
numbers that have already been measured, but to learn something new. It is
often possible to learn things from "negative" results, but it takes time
and effort.
Be sure to visualize your trajectories to make sure you are sampling
configurations that look reasonable. Make sure the binding affinities are
in a reasonable range. Look at each of indidivual components of the binding
energy to see if anything stands out there.
...good luck....dac
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Apr 12 2021 - 06:30:04 PDT
