From chemistry-request -A_T- ccl.net Mon Mar 23 12:32:06 1992 Date: Mon, 23 Mar 1992 10:48 EST From: CHIRLIAN \\at// CC.BRYNMAWR.EDU Subject: Batchmin using the GB/SA solvent model To: chemistry;at;ccl.net Status: R I have a question for anyone who is using BatchMin (part of MacroModel) to do Monte Carlo conformational searches using the GB/SA solvation model. When I reminimize the structures that are obtained in the Monte Carlo search, the energies _go up_ significantly (~12-13 kJ). The manual refers to a minor probelm invoving the effective radii which may result in larger gradients (but makes no mention of changes in energy) and suggests using the CHGF command to produce a charge file which can be used for subsequent minimizations. I have not been able to get this command to work, the output always includes the lines: CHG file not read, old charges retained and Charges come from force field. I am using the same command files as in the manual except I am using the MM2 force field rather than the OPLSA. I would appreciate hearing from anyone who has experience with these types of calculations. Specifically I would like to know if anyone else has had the energy rise when reminimizing structures (note: this only happens when solvent is included). If the energies rose, did writing out the .chg file solve the problem? How were you able to get the .chg file written, rather than having the charges read from the force field. Any insight would be appreciated. I am not concerned about the absolute energies, but would like to know if the ordering of structures can be trusted. I would also like to know why the energy _rises_. Thanks in advance for any help. Please e-mail any responses directly to me and I will summarize if there is sufficient interest. Lisa Chirlian