The previous minimization calculation relaxed water around the solute molecule. If we want to be really meticulous, we can now do the reverse and minimize only the solute while holding the water molecules fixed, and then relax the whole system together. However, keeping in mind that our goal is just to remove bad contacts, there is no need to go overboard with minimization. We will proceed directly to minimizing the entire system as a whole:
Again, save this input in a file and name it something meaningful like "min_all.in", and feed it to sander.
Side Note: This reminds me to tell you that while carrying out a simulation, the number of files can quickly add up. If you have a habit of keeping a separate notebook to keep track of things, that's great, but if you are like me and rarely writes down any thing on paper, it is very important that you have a good naming system for the files you create. Having a good naming system can make your work more or less self-documenting whereas having a bad naming system can result in a lot of unnecessary headaches a few weeks down the road.
This will take another 5 ~ 10 minutes on a fast Pentium IV machine, so it's time to take another short break.
When you come back, don't forget to check your results by generating a pdb file from the restart file just like we did before.