Conformational pool, potential conformers

A method called PARSE (Probability Assessment via Relaxation rates of a Structural Ensemble) is described for determination of ensembles of structures from NMR data. The problem is approached in two separate steps (1) generation of a pool of potential conformers, and (2) determination of the conformers probabilities which best account for the experimental data. The probabilities are calculated by a global constrained optimization of a quadratic objective function measuring the agreement between observed NMR parameters and those calculated for the ensemble. The performance of the method is tested on synthetic data sets simulated for various structural ensembles of the complementary dinucleotide d(CA) d(TG). 

Theory. We will assume now that a pool of n potential conformers has been... 

PDQPRO Results. The PDQPRO program was run separately for the "experimental" rates of each of the eleven two-state B-A equilibria, using a pool of 402 potential conformers. The goal was to pick two correct structures out of 402, and to predict their probabilities. The PDQPRO was successful in all eleven cases, selecting the correct structures (B- and A-forms) and calculating the correct probabilities within the precision of the computer digital representation ... 

In conclusion, we have shown that correct conformers and their probabilities can be retrieved with the PDQPRO program once they are present in the pool of potential conformations. However, the problem of adequate conformational sampling still remains not completely solved. 

---