I want to perform MD for protein-ligand but I have not facilities, I tried with GROMACS web server and Amber web server, but its access is only limited to some countries.
For a simulation about 1 ns using Gromacs, I believe you could use the PCs you have.
For example, I have an old ACER (produced in 2009), with 4 CPU cores, 4GB RAM and 500 GB hard disk. Then I installed Ubuntu 14 on the computer and installed Gromacs also. In case you need to simulate a protein-ligand complex, AmberTools (free for academic users) may also be requried. By making use of any PCs you have, you may run simulations on these PCs. If you don't want to use Ubuntu OS, you may also install Gromacs on Windows (together with Cygwin). However, a linux OS is preferred.
And from my understanding, an online server for protein-ligand complex simulation is not available yet, since the topology for the small ligand is not that straight forward. So we need to prepare the ligand topology at our own side, not the server side. And a simulation about 1 ns would not require much computational resources.
Another option is, you may find someone really know MD simulations and ask him/her to assist you or cooperate with you.
What is free can sometimes be very expensive. Investing into few CPUs will have many benefits that will more than offset the initial (and maintenance) costs; most notably: unlimited access, total control, seamless workflow.
For a simulation about 1 ns using Gromacs, I believe you could use the PCs you have.
For example, I have an old ACER (produced in 2009), with 4 CPU cores, 4GB RAM and 500 GB hard disk. Then I installed Ubuntu 14 on the computer and installed Gromacs also. In case you need to simulate a protein-ligand complex, AmberTools (free for academic users) may also be requried. By making use of any PCs you have, you may run simulations on these PCs. If you don't want to use Ubuntu OS, you may also install Gromacs on Windows (together with Cygwin). However, a linux OS is preferred.
And from my understanding, an online server for protein-ligand complex simulation is not available yet, since the topology for the small ligand is not that straight forward. So we need to prepare the ligand topology at our own side, not the server side. And a simulation about 1 ns would not require much computational resources.
Another option is, you may find someone really know MD simulations and ask him/her to assist you or cooperate with you.
Formerly, Haddock website ( http://haddock.science.uu.nl/enmr/services/GROMACS/main.php ) enabled users to run some limited MD simulations but currently that server is closed due to having technical reasons. However, you can run your MD simulations by using GPU-based operating systems either through Gromacs tool or NAMD package.
Totally agree with Martin, cheap is expensive, although doing a dynamic simulations seems somewhat simple there is a high degree of complexity that increases with the type of system and the number of molecules included in a complex. The fact that an online server throws you results quickly does not mean that they are good or they have any meaning, hard and pure dynamics is the best.