| 概要： |
Considerable attention has been paid to the functions of RNAs, especially those of regulatory non-coding RNAs.
It is widely believed that there is a strong correlation between the 3D structure of an RNA molecule and its function.
A set of base pairs is called a secondary structure, which shapes the substructure of the 3D structure.
Since experimental determination of RNA 3D structures is difficult and their structures are hierarchical, secondary structure prediction provides a major key to elucidating the potential functions of RNAs.
In the former part of the talk, we present IPknot, a novel computational method for predicting RNA secondary structures with pseudoknots based on maximizing expected accuracy of a predicted structure.
Pseudoknots, found in secondary structures of a number of functional RNAs, play various roles in biological processes.
The problem of maximizing expected accuracy is solved by using integer programming with threshold cut.
IPknot is validated through extensive experiments on various data sets, indicating that IPknot achieves sufficiently better prediction accuracy and faster running time as compared with several competitive prediction methods.
The latter part of the talk focuses on predicting RNA-RNA interaction, leading to identifying possible targets of non-coding small RNAs that regulate gene expression post-transcriptionally.
We present RactIP, a fast and accurate prediction method for RNA-RNA interaction of general type using integer programming with threshold cut similar to the methodology of IPknot.
Experimental results on real interaction data show that prediction accuracy of RactIP is at least comparable to that of several state-of-the-art methods for RNA-RNA interaction prediction.
Moreover, we demonstrate that RactIP can run incomparably faster than competitive methods for predicting joint secondary structures.
(The talk is in Japanese)
|