peptide tends to migrate out the water box

From: Francesco Pietra (
Date: Thu Apr 17 2014 - 17:02:01 CDT

During accelerated MD (boost on dihedral) in TIP3wader periodic box with
namd2.9_CUDA, the peptide is folding correctly, albeit still at an
incomplete stage. However, after barely 500,000 steps ts=2fs, the peptide
has gradually migrated toward a face of the box.

I never observed before such a behavior for a box built with "padding 15",
i.e., with plenty of water around the peptide under non-CUDA accelMD for
these types of peptides.Flags that might be relevant to the phenomenon

rigidBonds all ; # needed for 2fs/step
# rigidBonds water
rigidTolerance 0.000001
exclude scaled1-4
cutoff 12.0
switching on
switchdist 10.0
pairlistdist 13.5 ;# cutoff +3.5
margin 0
1-4scaling 1.0
PME yes

wrapNearest no
wrapAll on

Thanks for advice.

francesco pietra

This archive was generated by hypermail 2.1.6 : Thu Dec 31 2015 - 23:20:44 CST