RNA plays key roles in many biological processes, and its function depends largely on its three-d... more RNA plays key roles in many biological processes, and its function depends largely on its three-dimensional structure. We describe a comparative approach to learning biologically important RNA structures, including those that are not the predicted minimum free energy (MFE) structure. Our approach identifies the greatest conserved structure(s) in a set of RNA sequences, even in the presence of sequences that have no conserved features. We convert RNA structures to a graph representation (XIOS RNA graph) that includes pseudoknots, and mutually exclusive structures, thereby simultaneously representing ensembles of RNA structures. By modifying existing algorithms for maximal subgraph isomorphism, we can identify the similar portions of the graphs and integrate this with MFE structure prediction tools to identify biologically relevant near-MFE conserved structures.
RNA plays key roles in many biological processes, and its function depends largely on its three-d... more RNA plays key roles in many biological processes, and its function depends largely on its three-dimensional structure. We describe a comparative approach to learning biologically important RNA structures, including those that are not the predicted minimum free energy (MFE) structure. Our approach identifies the greatest conserved structure(s) in a set of RNA sequences, even in the presence of sequences that have no conserved features. We convert RNA structures to a graph representation (XIOS RNA graph) that includes pseudoknots, and mutually exclusive structures, thereby simultaneously representing ensembles of RNA structures. By modifying existing algorithms for maximal subgraph isomorphism, we can identify the similar portions of the graphs and integrate this with MFE structure prediction tools to identify biologically relevant near-MFE conserved structures.
Uploads
Papers by Kejie Li