![]() |
|
![]() |
|
Single Stage - The default stages of analysis can be invoked individually (normally not done) or in sequence by running "Default Execution". "Default Execution" is a programmable macro that normally is set to run two iterations of each analysis stage.
- Create Ladders - specify unambiguous chemical shifts of all spin systems; for spin systems with highly degenerate root frequencies, large numbers of noise peaks, or intra-residue peaks which cannot be easily distinguished from sequentials, chemical shifts may be left undesignated.
- Match Unambiguous - simple, iterative matching is done between all CO- and CA ladders to establish links between GSs. (See also Match Cycle)
- Match Degenerate - undesignated intra-residue chemical shifts are reconsidered, and peaks which were previously deemed "sequential" peaks detected in an "intraresidue" experiment are used to fill out incomplete ladder specifications. Highly degenerate root frequencies and/or large numbers of noise peaks may still cause some shifts to remain undesignated. Simple, iterative matching is then performed as in the previous stage.
- Extend Segments- as noted, the previous stages may leave many chemical shifts undesignated due to degenerate spin system roots and/or noise peaks. In this stage, the N- and C-terminal spin systems of assigned segments are used to form "templates" which specify the expected chemical shifts of the following or preceding spin systems. These templates are used to search the peaks associated with unassigned spin systems to see if a good match to the templates can be uniquely constructed. If so, then the chemical shifts on the corresponding ladders are designated, and the links and assignments are established. The user can specify the set of unassigned spin systems to be considered, which may include weaker spin systems and/or side chain spin systems if desired. In the default batch job, only side chains are excluded.
- Match Weaker GSs - weaker (unassigned) spin systems are restored to the general list of unassigned spin systems and simple iterative matching is executed to establish links and assignments.
- Match Inconsistent - (Proposed - not yet implemented) A number of promising matches may have been found in earlier stages which were rejected due to inferred type incompatibilities resulting from incorrectly designated shifts. In this stage an attempt is made to identify the specific shifts which caused the type violations and to substitute these with suitable values from the other peaks available to those spin system so as to obtain type compatibilities with unassigned residues in the sequence.
- Reason By Elimination - (Proposed - not yet implemented) Occasionally near the end of analysis, only a handful of isolated residues remain unassigned, and a one-to-one mapping of residues to spin systems can be established by considering only spin sytem type probabilities. As always however, spin systems assigned adjacent to one another in the sequence should obtain a reasonable match between the associated CA- and CO-ladders.
- Finish Assignments - This will consist of a "bag" of methods currently being explored to finish out the assignments.
- Match Cycle - given a starting threshold, step size, number of iterations, and list of match dimensions, iteratively establish links by decrementing the threshold and/or the number of match dimensions on each iteration
- Incremental Match - given a starting threshold, step size, and list of match dimensions, iteratively establish links by decrementing the threshold only on each iteration
- Constrain Segments - constrain the possible assignments of an unassigned segment of of linked spin systems by possible (1) links and/or (2) types.
- Update Probabilities - update the probability scores derived from the designated chemical shifts by removing probable types which do not agree with the list of possible assignments (if defined) for that spin system. A spin system with no established links will have NO possible assignments yet, and will be unaffected by this action.
![]() |
|
![]() |
|
![]() |
|
![]() |
|
1. An ideal GS is created for each non-proline residue whose CO- and CA-ladders have only the expected Ca and Cb shifts for the preceding and current residues respectively. These ideal spin systems are then examined to determine their theoretical possible assignments, and the total number of possible GS assignments for all assignable residues is computed.For example, consider a sequence of five identical residues: AAAAA. Here, all but the first residue are assignable, and each position has four possible GS assignments. Thus N = 4 + 4 + 4 + 4 = 16, and the average is just 16/4 = 4. Similarly, given a sequence of ten alanines, we would have N = 36, with an average of 9 possible assignments per residue. In both cases however, dividing again by n yields a value of 1, which is the theoretical maximum sequence degeneracy. More generally, the final sequence degeneracy computed this way will always fall between 0 and 1, and provides a rough measure of redundancy in the sequence independent of n.2. This total number of possible assignments (N) is then divided by n to obtain the average number of possible assignments per residue (N/n).
3. To obtain a measure of sequence degeneracy which is independent of sequence length, the average obtained in the previous step is again divided by n.
- Link Discrepancies - analyzed mapped segments for descrepancies in linking. Each pair of adjacently assigned GSs has an associated link score, which in a reasonably good data set should be greater than 0.9. Lower values may indicate an error in the sequential assignments or designated chemical shifts.
- Type Discrepancies - analyze the typing of each spin system. Each assigned spin system has an associated type probability score for the residue to which it is assigned. Values above 0.1 are reasonable, as even an exact match to the expected values may yield a type probability of less than 0.5, due to the overlap in expected values and standard deviations of residue types. Values below 0.1 do not necessarily indicate errors, as the observed shifts are sensitive to conformation. Accordingly, the recommended threshold for checking type discrepancies is 0.1 or less.
- Unexplained Peaks - display peaks that have not been used in any resonance assignment.
- Assignments - display assignment statistics.
![]() |
|
![]() |
|