Any solution (old or new) that I reloads appears with >-250000 ideality score for each ligand.
Same if I change track and come back to the new track.
To reproduce:
reset puzzle (points about 8990 pts), create new track B6, return to default track, return to B6: about -490 000 pts (even after restore credit best)
I think I solved this, but there are some caveats. The solution you shared with me is definitely messed up. I changed a couple of things to loading solutions, so this may have fixed it. I cannot reproduce the error now; however, these changes do not work for your previously saved solution; that may never be fixed.
Just to make sure that I did this right before posting an update:
1) Open a saved puzzle (in this case yours)
2) Reset puzzle
3) Modify ligand (substitutions, adding fragments, wiggle/shake/mmFF)
4) Save solution as test1
5) Open another solution and/or reset puzzle
6) modify ligand again
7) Do not save, but load previous solution test1
When I do this, I get the same score as before.
It looks like the NADPH molecule was not loading correctly. Can you load up your solution again and show hydrogens for the molecule? For me, when I load your solution it shows impossible bonds for the NADPH molecule. Is that what you get too?
Thanks Bruno,
I have been able to reproduce this problem and I believe I have a solution. Now, instead of crashing, two things will happen. The protein will "explode", meaning that in some cases, wiggle cannot resolve the clashes introduced from design and the protein will open up to accommodate the changes. The second thing that can happen is the ligand will move out of the binding pocket. What happens depends on the severity of the clashes and only happens in extreme cases. Most of the time things can be resolved.
If I use the idealize tool on the NADPH molecule (the one you cant design), it fixes the molecule and the score goes back up to the high score for the puzzle. This works for all the puzzles shared with me that had bad scores when loaded. I believe the upcoming patch will no longer have these errors when a puzzle is saved.
Hi Berto,
I moved MMFF minimization to the submenu of wiggle/shake/ligand design. I did this because its more in line with the philosophy of what wiggle/shake do. I also added the stop icon to the button so that you can press it instead of the upper left hand corner button.
As for the long stopping session…This is a little harder thing to control. Right now, for MMFF, there are 25 rounds of 10 steps of minimization. At the end of each set of 25 rounds, it increments the number of rounds. I have played with various rounds/steps, but 25 seems to work the best. In the future, I believe I can add in the option to change this, but for now, this is what works best.