[gmx-users] obtaining series of data from same run and gen-vel

[Felix Y Yang]

Hi,


I'm trying to repeat a 5 nanosecond protein md run in hopes of getting multiple different trajectories to analyze, and I've read around the forum that to do that I change my parameters to gen-vel = yes, and continuation = no.


Right now, I'm using the same NVT output file from the prior step (equilibrated at 300K) and changing my final NVE step parameters from gen-vel = no to gen-vel = yes, and continuation = yes to continuation = no to get my md trajectory. I do end up with different trajectories each time, but I'm wondering if it is appropriate to use gen-vel = yes when I don't even have a thermostat on?


Should I just change gen-vel = yes in the NVT step right before NVE instead?


Also if I'm heating up a few residues in the protein to a higher temperature than 300K, while the rest of the system stays at 300K, how would I set the gen-temp appropriately with gen-vel = yes?


Thanks,


Felix