Dear Alan, This depends very much on whether the structures are similar in the helical conformations or not, and on what question you are trying to address. If these are nearly identical structures, it really does not matter much what atoms you use for superposition, as long as they lie along the backbone. Indeed, for general superpositions, you could use all the atoms of the backbone. The most common single atom type would be the C1' carbons of the sugar, since the C1' to C1' distances should be constant across Watson-Crick paired bases and should be similar along the individual strands regardless of the sequence. This would allow you to compare the helical core of the structures without worrying about variations along the backbone, such as sugar puckers, phosphate orientations, etc.
However, if you are comparing across conformations (e.g., between A- versus B- type helices), there would not be many atoms that would be highly superimposable, in which case a general all atom superpositioning would probably be most appropriate (although I am not sure what you would learn from that other than general orientation). ------------------------------------------------ P. Shing Ho, Ph.D. Professor & Chair Biochemistry & Molecular Biology 1870 Campus Delivery Colorado State University Fort Collins, CO 80523-1870 970-491-0569 (phone) On 1/22/13 8:00 AM, "Alan Cheung" <che...@lmb.uni-muenchen.de> wrote: >Dear all - is there a convention for superposing nucleic acids duplexes >of unrelated sequence? > >i.e. which atoms should be superposed : C4' of the sugar, P of the >backbone, O3'/O5' of the backbone, or some combination thereof? > >Alan > > >-- >Alan Cheung >Gene Center >Ludwig-Maximilians-University >Feodor-Lynen-Str. 25 >81377 Munich >Germany >Phone: +49-89-2180-76845 >Fax: +49-89-2180-76999 >E-mail: che...@lmb.uni-muenchen.de
