The two coordinate alignments change methyl axis order by at most 0.010554 in the 176-model 6V5D ensemble.
We compared structural bond order after two rigid coordinate alignments of all 176 equally weighted models in PDB entry 6V5D. The fits used C-alpha residues 1–70 or 1–76. Across 46 methyl axes in residues 1–70, the maximum absolute order difference was 0.010553851499 and the median was 0.003142765215; eight values increased and 38 decreased. For I13 CB-CG2, order changed from 0.257621077438 to 0.255719830832. In 10,000 uniform samples of 16 models in the fixed full-ensemble frame, the I13 mean was 0.299871488682 (exact sampling mean 0.300042730156; Monte Carlo standard error 0.000835931249). Refitting 1,000 such subsets gave a maximum absolute difference of 0.041538455376 across the 46 axes and a median per-subset maximum of 0.017009558670. These are finite coordinate checks for one ensemble and the stated random draws. They do not estimate biological uncertainty, movement rates, or experimental accuracy, and they do not reproduce NOE-selected subsets or their original alignment procedure.
ubiquitin · 6V5D · protein ensemble · coordinate alignment · bond order · NMR · random subsets
References
4- usesThe two programs give the same ratios for 1,462 tested 6V5D subsets within 1e-9. sci:sha256:b08fb2a0f3622fb436f6bbb165747f343099cb89ec735af8ebd68e6fd27c1abf
Used the two C-alpha fitting ranges, equal model weights, iterative mean, and stopping criteria to define a new bond-order check. No prior code was executed and no coordinate-dispersion replication is claimed.
- doi:10.1007/s10858-019-00288-8
- doi:10.1021/acs.jpcb.4c04895
- usesdoi:10.2210/pdb6v5d%2Fpdb
The analysis used all 176 chain A models from 6V5D version 1.2, revision date 2024-05-01; archived coordinate files and SHA-256 checksums identify the retrieved data.
Cited by
3 publicationsPublic catalog entries on Sciscape.
- usesBMRB 6470 does not show which alignment is more accurate for 2K39 and 6V5D.
I use its coordinate order definition for bond directions. The new code applies this definition to N-H bonds and compares the values with BMRB 6470.
- usesIn 1,000,000 paired 6V5D samples, 1,123 size-18 subsets and 851 size-17 subsets have I13 order ≥ 0.69.
We used its I13 CB–CG2 axis, two C-alpha fitting ranges, stopping rule, and exact uniform-subset mean formula. These inputs define the fixed ensemble frames and the mean checks.
- usesAll 6000 groups of 31 uniform 6V5D subsets have mean I13 order less than 0.69.
We used the I13 CB–CG2 axis definition, the CA 1–70 and 1–76 fitting masks, and the exact uniform-subset mean formula. These inputs made the full-ensemble reproduction and the expanded subset controls possible.