RNA in the Evolutionary Context

The lab of Professor Lauren A. Meyers in the Department of Integrative Biology at UT Austin uses TACC resources to study the evolution of RNA molecules. RNA (ribonucleic acid) is one of the building blocks of life and a key molecule used in reconstructing evolutionary history. In particular, they perform computationally intensive in silico selection of RNAs on the basis of their folded configurations. That is, they consider virtual populations of RNA molecules, make thermodynamically based predictions of their structures, and assign fitness scores to the molecules based on the similarity of their structures to an ideal structure. The fittest are replicated preferentially with the occasional mutation.

In one project, Meyers and graduate student Matt Cowperthwaite are studying evolutionary dynamics under heterogeneous environmental conditions, an important and general scientific problem considering that most of life has evolved under changing conditions. In their simulations, they periodically alter the ideal structure and analyze the traits, environmental stability, and variability found within the evolving populations.

In a second project, they study the nature of the RNA "fitness landscape." The fitness landscape is a metaphor first used by Sewall Wright in 1932 to describe the relationship between genotypes and their fitness. Genes underlie the traits (phenotypes) that in turn impact the fitness of an organism. The map from genotype to phenotype can be quite complex, and very little is known about biologically realistic maps of this kind. Thus Meyers and her students are using RNA to gain such insight. In particular, the folding of an RNA from primary sequence into secondary structure is a map from genotype to phenotype that underlies the function and thus the fitness of a molecule. By computationally predicting the structures of a vast array of RNA molecules, they are studying properties of the sequence-structure map, including the distribution of local optima, the shape correlations among genetically similar sequences, and a number of other statistical features that provide insight into and perhaps the ability to predict RNA evolution.