Dear Dr. Sali & colleagues,

I used Modeller6v2 to build 3-D models of some unsolved protein sequences, using as input sequence alignments obtained from the threading program, 123D. Since Modeller performs a partial rfefinement of the structure it builds, do you think it is possible that the fold of the final 3-D structure would be different from the fold of the alignment template found by 123D? Have you seen a case where the fold was altered by the refinement performed by Modeller? How often does this happen and in what particular cases?

Thank you very much for your time and your help.

Very sincerely yours,

V. Reyes

====================================================== >>>>>>>>>>>> VICENTE M. REYES, Ph.D. <<<<<<<<<<<< Dept. of Pharmacol., Sch. of Pharm. & Pharm. Sci., & Integrative Biosci. Dept., S.D. Supercomputer Center Univ. of California-San Diego, La Jolla, CA 92093-0505