I have a protein homology model which I am currently running MD simulations for.
I am asking if there is a way to force this protein to adopt a certain conformation of another homologous protein based on sequence alignment. I am currently running an MD simulation with NAMD. I am looking for a method to minimize the RMSD between the two proteins. I searched the NAMD documentation but could not find something similar. Is there is a way to do this in NAMD.?
Wouldn't it be much easier and efficent to perform simple structural alignment? I don't see the point to employ MD in this case...
Thanks Rafal for your answer..i did that at the beginning but i have observed that the model is slightly drifting gradually while the simulation is running...I think a gradual alignment by minimizing the RMSD while the simulation is running is more beneficial...
Actually why do you want to force it to setten RMSD?
Please remember that single RMSD value can be represented by many states of the system, so using this kind of constraint may be not enough.
Thank you for an interesting thought. I think it would be useful to restrain the distances between atoms of independent models. This is not RMSD, but may be used for the same purpose. Then, one of the models can be used as a fixed pattern or vice versa do dynamics simultaneously. I can introduce such restrains in Abalone. Are you interested in?
http://www.biomolecular-modeling.com/Abalone/index.html
Thanks Marko for your suggestion. Indeed i had recently looked into the colvar module and read the paper of the authors, it does not have something similar...the ultimate goal is a tool to minimize the RMSD based on sequence alignment..
Thanks Alexey for your nice interest..I had a look on the Abalone site and was really surprised by the awesome capabilities and interesting alghorithmic features..i am not sure about the capabilities of Abalone for MD for such big systems (>500,000 atoms)..You can give a try and i can as well...Even if i do not use it myself, i think it will be a very nice add for your software, it will be something that no body else did..when you got, a Nature paper for this, do not forget to cite this researchgate question..:)
No. 500,000 is not possible. This is generally not possible on the PC. I plan to expand opportunities to approximately 50 000 - 100 000 but not more.
However, it seems to me that this size model in any case need to be considered in the coarse grain. After all, even for a small model the trajectory shorter than a nanosecond of little use. For millions of atoms, it will be probably a microsecond.
- Badges
- Science topic