Algorithms for biological pathway inference and RNA secondary structure analysis
MetadataShow full item record
Biological systems often demonstrate sophisticated structure features that are important to understand functions of living organisms. Researchers in computational biology have been interested in two typical types of structures. One is about intrinsic relationships among biological objects like genes, transcriptional factors, RNAs, operons, etc. The other is about the physical structures that biopolymers can fold into, such as protein tertiary structures and RNA secondary structures. This dissertation investigates the modeling and prediction of these two types of structures in general and presents case studies in solving a few speci c problems in computational biology involving such structures. In particular, it presents algorithms for comparative biological pathway inference, ab initio RNA secondary structure prediction, and learning of stochastic grammar models for RNA pseudoknots, based on e ective models built upon advanced graph theory and stochastic grammars. The algorithms, via sophisticated computation techniques, practically cope with the computational intractability inherent of these biological problems.