Statistical mechanics of RNA folding: importance of alphabet size. Author Ranjan Mukhopadhyay, Eldon Emberly, Chao Tang, Ned Wingreen Publication Year 2003 Type Journal Article Abstract We construct a base-stacking model of RNA secondary-structure formation and use it to study the mapping from sequence to structure. There are strong, qualitative differences between two-letter and four- or six-letter alphabets. With only two kinds of bases, most sequences have many alternative folding configurations and are consequently thermally unstable. Stable ground states are found only for a small set of structures of high designability, i.e., total number of associated sequences. In contrast, sequences made from four bases, as found in nature, or six bases have far fewer competing folding configurations, resulting in a much greater average stability of the ground state. Keywords Structure-Activity Relationship, Base Sequence, Molecular Sequence Data, Models, Statistical, Models, Molecular, Computer Simulation, Reproducibility of Results, Base Composition, Nucleic Acid Conformation, Sensitivity and Specificity, Sequence Analysis, RNA Journal Phys Rev E Stat Nonlin Soft Matter Phys Volume 68 Issue 4 Pt 1 Pages 041904 Date Published 10/2003 Alternate Journal Phys Rev E Stat Nonlin Soft Matter Phys Google ScholarBibTeXEndNote X3 XML