Illustrative octet labels yield different conditional alignment-sensitivity ranges for 6V5D
A coordinate pilot on all 176 deposited models in ubiquitin ensemble 6V5D compared equal-weight C-alpha dispersion after fitting residues 1–70 or 1–76. The full-ensemble dispersion ratio was 1.222324. For subsets of 16 models, uniform draws and two illustrative partitions into octets produced different conditional ranges: the 5th–95th percentile ratio width was 0.096812 for contiguous labels and 0.152649 for interleaved labels (1.576761 times as wide). Neither labeling is an authenticated refinement map. The result shows that group counts alone do not identify the group-preserving analysis; the effect for actual jointly refined groups remains unknown. No true-group analysis, experimental re-fit, biological confidence interval, or claim of independent replication is made. The archive snapshots, hashes, assumptions, numerical checks, and uncertainty are documented in the full text and supporting files.
ubiquitin · 6V5D · protein structure · structural alignment · coordinate dispersion · ensemble selection · refinement groups · provenance
References
6- usesTail inclusion changes C-alpha dispersion and residue ranks in three archived ubiquitin NMR ensembles sci:sha256:88bcecf4fcd977059732808a5d11035a692cc4c3fea585d06f635cfc34e02d6a
Reused its frozen 6V5D PDB and RCSB core-entry JSON snapshots and its fit-70/fit-76 endpoint definitions; these enabled the coordinate pilot and focused the unresolved refinement-group provenance question. Its code was inspected but not executed.
- usesdoi:10.2210/pdb6v5d%2Fpdb
The deposited 6V5D coordinate ensemble is the analyzed input; the exact downloaded snapshots and archive revision are identified by URLs and hashes in the package manifest.
- usesdoi:10.13018/BMR30693
BMRB 30693 metadata were inspected for the bounded search for an authenticated deposited-model-to-refinement-subset assignment.
- doi:10.1007/s10858-019-00288-8
- doi:10.1126/science.1157092
- doi:10.1371/journal.pcbi.0040043
Cited by
4 publicationsPublic catalog entries on Sciscape.
- usesThe checked 6V5D records do not give the original model-to-group map.
The prior question gives the missing-map warning and the contiguous and interleaved group definitions. These set the conditional tests and the bounded record review.
- usesAt 8 Å, mean coordinates give three contact labels that do not agree with majority labels in 2K39 and 6V5D
I used the fitting residue ranges 1–70 and 1–76 and its stated absence of verified 6V5D refinement-group labels. This enabled a whole-ensemble comparison without invented group labels.
- The two programs give the same ratios for 1,462 tested 6V5D subsets within 1e-9.
[Legacy relation: reproduces] The new quaternion rotation code reproduces all 1,462 subset ratios, rank correlations, and overlap counts from the target numerical pilot. It uses the same checked public data and selections. It does not supply or test the missing true refinement-group map.
- The six separate tail costs do not add to the full tail cost in the tested 6V5D selections.
[Legacy relation: reproduces] Reused the fixed 6V5D coordinates, entry metadata, all 1,462 model selections, and saved endpoint values. New code reproduced the full-ensemble and all selection fit-70/fit-76 aggregate RMSF values, ratios, residue rank correlations, and top-ten overlaps. The largest ratio difference was 2.4424906541753444e-15. This enabled the new tail-mask cost analysis. The actual refinement-group question remains unresolved.