I wish to do some molecular docking using AutoDock4.2 softaware. But I do not know how to calculate the free energy of binding of a complex. Can anyone suggest me a way to do the free energy calculations using any software?
Dear Renjith,
I think a Comparative Binding Energy (COMBINE) Analysis can be useful. You can find more information and software on the link below.
All the best,
Miguel
http://farmamol.uah.es/index.php/en/2-uncategorised/4-gcombine
I have used MM/PBSA in AMBER to calculate binding free energy. The tutorial is at http://ambermd.org/tutorials/advanced/tutorial3/. I recommend you use AMBERTOOLS14 for this calculation and pay attention to the internal dielectric constant and entropy, which will dramatically affect your results.
Most docking programs will provide you with an estimation of the binding energy. In order to do so, they use different kinds of so called "scoring functions". These are quick and dirty methods. There are different kinds of scoring functions, you can easily find literature on the subject. The one used by AutoDock, for instance, is force field-based and adjusted according to empirical data.
Now, in most instances, the values provided by those scoring functions will be largely inaccurate. You can use different scoring functions and analyse the results independently or combine them with so called "meta-scores" such as that indicated above. If you have empirical data on complexes that are similar to the ones you are studying, you can have an indication how well each method will perform in your case.
If you need more accuracy you can go for MM/PBSA or MM/GBSA which typically involve the use of molecular dynamics simulations and are therefore computationally expensive. There are methods, however, to carry out those kinds of evaluations from conformational ensembles generated without the need of MD simulations.
Yet more accurate and expensive would be to carry out Free Energy Perturbation (FEP) calculations.
Even more expensive methods making use of different QM levels of theory include QM/MM-GBSA, QM/MM-PBSA, MM-QMSA and others.
The choice of method will depend on the accuracy you need, the time and computational resources you have available and, related to this, the size of your problem.
I've used Xscore for this purpose.
It's free...but you need to get license for using it.
http://sw16.im.med.umich.edu/software/xtool/
I think its better to go with AutoDock Vina generated binding energy as it follows X-Score like scoring function (see http://onlinelibrary.wiley.com/doi/10.1002/jcc.21334/full) . Vina article has cited X-Score saying that it follows X-Score function.
- Badges
- Science topic