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 merging the assignments from two projects A and B containing the assignments of two molecules (or two chains of a heterodimer, or homodimer)
gzipped tarfile with Screenshots of procedure
- 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.
- 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.
- Project ProteinA: Use the SequenceExplorer? to determine the maximum residue number in the Sequence and the maximum SpinId in the SpinList?.
- 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)
- Navigate to the SequenceExplorer? in Project B and renumber the 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).
- Load version 12 or later of WriteAssignments.lua into the repository containing Project B.
- Run version 12 or later of WriteAssignments.lua
- Select Project ProteinB
- checkbox
number by ChainId - Enter a number in the "Add offset to SpinId" field that is larger than the maximum SpinId determined in step 4 (e.g. 5000)
- Select format CARA.
- 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.
- Your WriteAssignment.lua dialog should look something like this (except you should set format to CARA)
- Duplicate project ProteinA (renaming it something like ProteinAandB).
- Project ProteinAandB: Merge the two projects:
- In the projects SequenceExplorer? assign the sequence of protein A the chain identifier A (right-click "set chain" and enter "A").
- 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".
- 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"
- 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.
- 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.