Dear Theresa, it depends what you consider as small. 50 residues are easy to do with NMR but at 200 it will take some time (maybe 1-2 years to complete an assignment and structure calculations). There are several difficulties in NMR e.g. to do the correct alignment of monomers in oligomeric complexes and the price of glucose which naturally limits the yield of your expression. Samples need to be stable, ideally at increased temperature. There are no reliable de novo (e.g. Arcimboldo) and molecular replacement approaches as in crystallography for larger proteins. Even very similar proteins or the same protein e.g. in different detergents may yield very different HSQC spectra. Typically proteins which yield nice NMR spectra are well folded and consequently also crystallize well and vice versa. You may do both in parallel or use the NMR sample afterwards for crystallization. Crystallization is typically much faster for structure determination. Best wishes Kornelius
On Sun, Jun 9, 2013 at 7:33 PM, Mark van Raaij <mjvanra...@cnb.csic.es>wrote: > Well, if you do NMR you avoid the possible bottlenecks of having to obtain > well-diffracting crystals, and having to phase the protein (i.e. obtain > SeMet protein crystals or suitable heavy atom derivatives; or a suitable MR > model). > But instead, you'll need to prepare labelled protein (15N and/or 13C), > which is expensive and for which your protein needs to be able to be > expressed in minimal medium, and your protein will need to be very soluble, > monodisperse (in general monomeric) and stable in a minimal NMR-compatible > buffer for data collections lasting for hours. Assigning all the protons > and calculating the final structure can also be months of work, while a > high-resolution crystal structure can be finished in days, if the > above-mentioned bottle-necks can be overcome. > > > On 9 Jun 2013, at 17:36, Theresa Hsu wrote: > > > Dear all > > > > A question for the cross-trained members of this forum - for small sized > proteins, is NMR better than crystallography in terms of data collection > (having crystals in the first place) and data processing? How about > membrane proteins? > > > > I would appreciate replies to the board, instead of off-board, to allow > for a good discussion. > > > > Thank you. > > > > Theresa > -- *Kornelius Zeth* *Email: kornelius.z...@gmail.com* *Unidad de Biofisica (CSIC-UPV/EHU) Barrio Sarriena s/n 48940, Leioa, Vizcaya* *SPAIN*
