Edit detail for StructureCalculationOfComplexes revision 17 of 40

Determining the structures of complexes or oligomers requires some additional procedures. Typically isotope-labeling of individual components helps to simplify the problem. The project is therefore divided into subprojects where only the resonance assignments of the isotope-labeled component are determined. Later the assignments of the individual components must be merged so that the NOEs between components can be analysed using experiments which use isotope-filtering and -editing to obtain only intermolecular NOEs.

CARA provides support for this analysis by allowing the definition of samples with different isotope-labeling patterns Moreover, SpinLinks? which represent NOEs? between spins can be defined to only appear in a subset of NOESY spectra. Via the SpinLink? visibility the subset of NOEs? visible in isotope-filtered/edited experiments can be represented. (See release notes for how to do this, further information is forthcoming on the wiki).

Here is a step-by-step procedure for determining the structure of a complex between two molecules A and B (or a heterodimer, or homodimer) Screenshots of procedure

1. Project ProteinA: assign protein A in a project using samples with labeled protein A and unlabeled protein B. In this project you will only have the sequence of protein A. You assign in the usual way for a single protein project.
2. Project ProteinB: assign protein B in a different project using samples with unlabeled protein A and labeled protein B. In this project you will only have the sequence of protein B. You assign in the usual way for a single protein project.
3. Project ProteinA: Use the SequenceExplorer? to determine the maximum residue number in the Sequence and the maximum SpinId? in the SpinList?.
4. Project ProteinB (note for a homodimer, create a copy of ProjectA naming it ProjectB. This can be used for chain B of the homodimer. Otherwise follow the steps below) a. SequenceTable?: renumber Chain of protein B starting with a number that is +1 relative to the maximum residue number of protein A in project A. Right-click on the first residue of protein B and select "renumber from here" entering the starting ChainNr?. (158 in the example shown here). b. Load the version 11 or later of WriteAssignments.lua into the repository containing Project B.
5. Run version 11 or later of WriteAssignments.lua a. Select Project ProteinB b. checkbox number by ChainId c. Enter a number in the "Add offset to SpinId?" field that is larger than the maximum SpinId? determined in step 4 (e.g. 5000) d. Select format CYANA. Your WriteAssignment?.lua dialog should look something like this e. Click OK (this writes out a sequence file starting at the ChainNr? of the first residue of protein B, and a chemical shift list .prot where each spin has an Id in the output file which is +5000 compared to the Id it has in Project B. We do this so that the sequence and chemical shift list can be merged with that of project A.
6. Duplicate project ProteinA (renaming it something like ProteinAandB).
7. Project ProteinAandB: Merge the two projects: a. In the projects SequenceExplorer? assign the sequence of protein A the chain identifier A (right-click "set chain" and enter "A"). b. Append the Sequence of protein B: right-click Append Chain, select the sequence file of protein B generated in step 5, and assign the ChainId? "B". c. read in the chemical shift list of protein B: right-click on the project ProteinAandB and "Import-Atom List" selecting the option AtomList? references "Residues"
8. Step 7 appends the sequence of protein B with its assigned chemical shifts to protein A. Next you can define two samples with either residues of protein A or residues of protein B 12C,14N-labeled.
9. Now load the filtered-edited NOESY spectra with either protein A or protein B 13C,15N-labeled and assign the corresponding samples to these spectra.