The domain common to both crystals constitutes the averaging mask. (you can make it around the pdb using ncsmask).
As for describing it to dmmulti, you need the rotation matrix which maps it from crystal 1 to crystal 2. The rotation matrix for crystal one is always the identity matrix. You will put the matrix mapping the domain from 1 to 2 as the rotation matrix for the second crystal. You may want to obtain a rotation matrix optimized against your current density first before supplying it to dmmulti (the rave utilities will do this for you). When you get to dmmulti, output the solvent masks and make sure they're accurate (or alternatively you can supply a solvent mask for each crystal separately). F On Sep 21, 2012, at 9:47 AM, Garboczi, David (NIH/NIAID) [E] wrote: > We are working with two crystals. Both crystals have a domain that is not in > the other. Both crystals have one domain that is the same in each. We'd > like the average the one domain between the crystals, but leave the other > domains unaveraged. > > How do we describe that to dmmulti? > thanks, > Dave > > -- --------------------------------------------- Francis E. Reyes PhD 215 UCB University of Colorado at Boulder