A. We developed and tested a protocol to reduce subjectivity from the process of ribonucleic acid (RNA) modeling. B. The modeling of specific RNA molecules has been continued. A. We have developed and extensively tested a modeling protocol to be applied with the computer program MC-SYM. The protocol is based on simple RNA structural principles such as base stacking and pairing. The protocol consists of a systematic search over the space of input scripts for the most restrictive one that produces solutions. This strategy allows to keep the number of solutions low and therefore simplifies a posteriori evaluation. The modeling of a transfer RNA (tRNA) molecule from structural data that was available prior to the crystal structure was used to evaluate the protocol quantitatively; by measuring the quality of the solutions produced by root mean square (rms) deviation from known crystal structure of tRNA-Phe and tRNA-Asp. The application of this protocol to tRNA allowed for the generation of the best model in almost all cases. B. The specific modeling of RNA molecules of biological interest is still in progress. These molecules are: a small metalloribozyme called leadzyme, group I introns, Ribonuclease P RNAs, and a spliceozyme. The leadzyme allowed us to study non-canonical base pairings and nucleotide binding to ions such as Pb2+ and Mg2+. The other molecules allow us to study the relative orientation and position of helices in large RNAs.
