Skip to main content
. 2012 Nov 23;55(1):33–46. doi: 10.1007/s10858-012-9683-9

Table 1.

Sequence variables and chemical shift corrections calculated by Pace regression

Atom const pre_a pre_c pre_g pre_gu pre_ug suc_a suc_c suc_g suc_gu
Part 1
A–H2 7.0299 0.6672 0.2521 0.6899 0.7637 0.2555 −0.5934 0.0658 0.2814 −0.3321
A–H8 8.1525 −0.4169 −0.1145 −0.3709 −0.3808 0.2107 −0.1006 0.0379 −0.0723 0.0458
G–H8 7.7036 −0.5622 −0.1595 −0.497 −0.5083 0.2215 −0.0383 0.0562 −0.0265 0.0369
C–H5 5.6724 −0.4611 −0.1625 −0.4304 −0.2815 0.0377 −0.0222 −0.0368 −0.0249 −0.032
U–H5 5.5639 −0.5133 −0.1778 −0.5018 −0.3641 −0.0469 0.0333 0.0165 0 0.0348
C–H6 7.8627 −0.3901 −0.0867 −0.2215 −0.2059 0.085 −0.0253 0.0363 −0.0288 −0.0833
U–H6 7.9946 −0.3643 −0.1083 −0.2523 −0.3392 0.0832 −0.0402 0.0332 −0.0555 −0.0809
A–H1′ 5.9905 −0.1042 −0.05 0 −0.0639 0 −0.0823 0 −0.0324 0.0528
G–H1′ 5.7539 −0.1822 −0.0584 −0.019 −0.068 −0.0373 0 0.0216 0.0198 0.0777
C–H1′ 5.56 −0.1805 −0.0825 −0.0446 −0.0492 0.039 −0.0226 −0.0139 0.0134 0.0711
U–H1′ 5.6203 −0.1679 −0.0865 −0.1049 −0.1992 0.0234 −0.0405 0.0097 0.0196 0.0571
A–H2′ 4.4575 −0.0269 0.0268 0.0309 −0.0499 0.1174 0.1041 0.0496 0.1613 0.1416
G–H2′ 4.4582 −0.0666 0.0389 0 0 0.0794 0.1367 0.0248 0.1286 0.0705
C–H2′ 4.3454 −0.1298 0.0188 0.0499 −0.0867 0.0363 0.1475 0 0.1557 0.1724
U–H2′ 4.4078 −0.0599 0.0404 0.0658 −0.0445 0.0618 0.0807 0.059 0.174 0.1431
A–H3′ 4.6788 −0.1275 −0.0229 −0.0794 −0.083 0.0436 −0.0377 0.006 0.0105 0.0593
G–H3′ 4.5076 −0.1292 −0.029 −0.0537 −0.1527 0.0644 0.0445 0.0459 0.0333 0.0674
C–H3′ 4.4794 −0.1336 −0.0189 −0.0431 0.0221 0.0364 0.0498 −0.0147 0.0454 −0.043
U–H3′ 4.495 −0.119 −0.0065 −0.0447 0.0458 0.0506 0.1024 0.0229 0.0657 0.0839
Atom suc_ug 5ter 3ter 3loop 5loop GU corr rms nobs xcorr xrms ntrim
Part 2
A–H2 0.0804 0.1558 0 0.0274 0.0622 0 0.9868 0.0562 162 0.9845 0.0608 1
A–H8 −0.0261 0.3206 0 −0.0175 0.08 0 0.9535 0.0575 157 0.9406 0.065 0
G–H8 −0.0671 0.3563 −0.0607 −0.0253 0.0332 −0.0529 0.9655 0.0658 288 0.9612 0.0697 2
C–H5 −0.0958 0.1245 −0.016 −0.0205 0.01 0 0.9674 0.0478 256 0.9635 0.0505 5
U–H5 0.047 0 0 0 0.0216 0.2747 0.9692 0.0537 178 0.9623 0.0594 0
C–H6 −0.1132 0.2326 −0.0359 −0.0428 0.0311 0 0.9332 0.0558 260 0.9249 0.0591 0
U–H6 −0.0457 0 −0.0287 −0.0341 0.0268 −0.0412 0.9444 0.0446 172 0.9331 0.0489 4
A–H1′ −0.0353 0.0966 0 −0.0173 0.0061 0 0.8476 0.0332 157 0.7905 0.0384 0
G–H1′ 0 0.1435 −0.0376 0 0 0 0.9278 0.0415 284 0.9209 0.0434 3
C–H1′ −0.0224 0.1351 0.0114 −0.0168 0.0121 0 0.8355 0.0427 253 0.7872 0.0482 5
U–H1′ 0.0328 0 0.0183 0.0145 0.0403 −0.0316 0.7411 0.053 172 0.6451 0.0609 1
A–H2′ 0.1044 0.0702 0 −0.0115 0.0291 0 0.7512 0.0618 143 0.6531 0.0714 0
G–H2′ 0.1022 0.074 −0.1122 −0.0326 −0.0257 0.1211 0.8675 0.0516 246 0.8359 0.057 2
C–H2′ 0.1198 0.085 −0.1316 0 0 0 0.9014 0.0589 217 0.8774 0.0654 4
U–H2′ 0.1691 0 −0.229 −0.0488 −0.03 −0.3095 0.92 0.0609 146 0.8999 0.0679 2
A–H3′ 0.0283 0.0235 0 −0.0248 0.0202 0 0.5997 0.064 129 0.4933 0.0708 0
G–H3′ −0.0887 0.1061 −0.0416 −0.0146 0.0244 −0.1527 0.7307 0.0807 226 0.6391 0.0914 3
C–H3′ −0.1182 0.1326 −0.0178 0.0096 −0.0224 0 0.7442 0.062 191 0.6942 0.0669 4
U–H3′ −0.0268 0 −0.0432 −0.0187 0.0159 0.0266 0.7047 0.0482 121 0.5928 0.0552 2

Output from the Pace Regression analysis. Each row represents an individual atom type in the specified nucleotide (e.g., A-H2 is the H2 proton of Adenine). The column labeled const represents the chemical shift of that atom in the triplet uXu when none of the additional attributes represented in subsequent columns are present. Contributions with values equal to 0 represent attributes that the Pace Regression algorithm found could not be supported by the data and were thereby automatically excluded from the regression analysis. The contribution from columns labeled pre_x and suc_x, where x is a,c,g, or gu are used where the preceding or succeeding nucleotide is not a u. A GU attribute represents the case where the nucleotide is in a GU, rather than GC, base pair, and can apply to the i − 1 (pre_gu), i + 1 (suc_gu) or central (GU) triplet (with the same approach used for UG wobbles). The 5ter attribute indicates the triplet is at the 5′ end (so there is no i − 2 nucleotide), and 3ter indicates the triplet is at the 3′ end (so there is no i + 2 nucleotide). The loop attributes indicate that the i − 2 (5loop) nucleotide or i + 2 (3loop) nucleotide is in a loop or mismatched base pair. The columns labeled corr and rms represent the correlation coefficient (corr) and the square root of the mean of squared deviations between predicted and experimental values (rms) for all the data in the fit. The columns labeled xcorr and xrms represent the same values, but calculated with 10-fold stratified cross-validation. The column labeled nobs represents the number of observations available and ntrim the number that were automatically eliminated as outliers