Modeling RNA Molecules

Abstract

A primary activity of scientific work is the construction of models to represent the nature and workings of phenomena we observe in the world around us. Models that represent the molecular components of living system in three dimensions (3D) and at atomic resolution are highly valued in molecular and structural biology. For example, the decipherment of the 3D structures of ribosomes, the complex protein-synthesizing nanomachines of the cell, represents a tremendous achievement, recently recognized with the Nobel Prize in Chemistry (http://nobelprize.org/nobel_prizes/chemistry/laureates/2009/). Nonetheless, this phenomenal success is tempered by the realization that even now, over 10 years after the first ribosome structures were solved, we still do not understand fully several aspects of their functioning. For all who have grappled with the complexities of ribosome structures, Richard Feynmann’s pithy statement, “What I cannot create, I do not understand,” rings especially true (Hawking 2001). This physics-based realization contrasts with another point of view of modeling. To paraphrase R. W. Hamming, who said, “The purpose of computing is insight, not numbers” (Hamming 1971), we should remember that the purpose of molecular modeling is functional insight, not detailed atomic models per se. Therefore, as we seek to improve our abilities to construct 3D models for molecules for which we do not yet have experimental atomic-resolution structures, we should bear in mind that it may not be necessary to achieve some arbitrary precision in the atomic coordinates to provide insight into biological function. Rather, we should think carefully to identify those predicted features that yield important insights (Table