new release

[Gary Huber]

Dear Browndye Users,

A new release is out.  

First of all, there have been some difficulties with initializing constraints, especially coplanar constraints, if they are not completely satisfied at the outset.  Browndye uses two algorithms to satisfy constraints.  One algorithm, which needs to be fast, relies on a previous, nearby configuration with constraints satisfied.  The other algorithm, which does not need to be fast, is used to satisfy the constraints at the outset of the simulation, and does not depend on a previous configuration.  This initial algorithm, which tries to iteratively satisfy the constraints, didn't always work.  It has now been replaced by a non-linear minimization algorithm which minimizes the deviation from the constraints. To the best of my knowledge, it should always work, as long as the initial chiralities are correct (i.e., it's not too far off).  When using constraints, care must be taken not to over-specify the constraints.  I'm working on some scripts to automate that process.

Second of all, we have included "dummy molecules" of non-interacting atoms, which can be used to define reaction criteria.  A dummy atom has no physical interactions, but moves rigidly with a core.  For example, a dummy copy of an atom of Molecule A could be attached to the binding site of Molecule B, and the reaction defined in terms of the distance between the dummy atom and its corresponding atom on Molecule A. (So far, I don't know how to best attach dummy atoms to chains.) It is described in the latest version of the manual.  

Thanks,

Gary