Edit detail for StructureCalculationOfComplexes revision 1 of 40

changed:
-
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:SpinLink 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).

 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: Determine the maximum ChainId 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 higher than the highest Chain number of protein A in project A. Right-click on the first residue of protein B and select "renumber from here" entering the starting ChainId number. (e.g. 200).
  b. Load the version 11 or later of WriteAssignments.lua into the repository containing Project B.

 5. Run version 11 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.
   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 Inter_A_lab_B_unlab 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.