Efficient parameter estimation for RNA secondary structure prediction.

@article{citeulike:1537640, abstract = {MOTIVATION: Accurate prediction of RNA secondary structure from the base sequence is an unsolved computational challenge. The accuracy of predictions made by free energy minimization is limited by the quality of the energy parameters in the underlying free energy model. The most widely used model, the Turner99 model, has hundreds of parameters, and so a robust parameter estimation scheme should efficiently handle large data sets with thousands of structures. Moreover, the estimation scheme should also be trained using available experimental free energy data in addition to structural data. RESULTS: In this work, we present constraint generation (CG), the first computational approach to RNA free energy parameter estimation that can be efficiently trained on large sets of structural as well as thermodynamic data. Our CG approach employs a novel iterative scheme, whereby the energy values are first computed as the solution to a constrained optimization problem. Then the newly computed energy parameters are used to update the constraints on the optimization function, so as to better optimize the energy parameters in the next iteration. Using our method on biologically sound data, we obtain revised parameters for the Turner99 energy model. We show that by using our new parameters, we obtain significant improvements in prediction accuracy over current state of-the-art methods. AVAILABILITY: Our CG implementation is available at http://www.rnasoft.ca/CG/ CONTACT: andrones@cs.ubc.ca.}, address = {Department of Computer Science, University of British Columbia, Vancouver BC V6T 1Z4, Canada and Department of Biochemistry \& Biophysics and Department of Biostatistics \& Computational Biology, University of Rochester Medical Center, Rochester NY 14642, USA.}, author = {Andronescu, Mirela and Condon, Anne and Hoos, Holger H H. and Mathews, David H H. and Murphy, Kevin P P. }, citeulike-article-id = {1537640}, doi = {http://dx.doi.org/10.1093/bioinformatics/btm223}, issn = {1460-2059}, journal = {Bioinformatics}, keywords = {estimation, parameter}, month = {July}, number = {13}, posted-at = {2007-12-19 19:44:23}, priority = {3}, title = {Efficient parameter estimation for RNA secondary structure prediction.}, url = {http://dx.doi.org/10.1093/bioinformatics/btm223}, volume = {23}, year = {2007} }

TY - JOUR ID - citeulike:1537640 L3 - citeulike-article-id:1537640 TI - Efficient parameter estimation for RNA secondary structure prediction. JF - Bioinformatics VL - 23 IS - 13 AD - Department of Computer Science, University of British Columbia, Vancouver BC V6T 1Z4, Canada and Department of Biochemistry & Biophysics and Department of Biostatistics & Computational Biology, University of Rochester Medical Center, Rochester NY 14642, USA. SN - 1460-2059 N2 - MOTIVATION: Accurate prediction of RNA secondary structure from the base sequence is an unsolved computational challenge. The accuracy of predictions made by free energy minimization is limited by the quality of the energy parameters in the underlying free energy model. The most widely used model, the Turner99 model, has hundreds of parameters, and so a robust parameter estimation scheme should efficiently handle large data sets with thousands of structures. Moreover, the estimation scheme should also be trained using available experimental free energy data in addition to structural data. RESULTS: In this work, we present constraint generation (CG), the first computational approach to RNA free energy parameter estimation that can be efficiently trained on large sets of structural as well as thermodynamic data. Our CG approach employs a novel iterative scheme, whereby the energy values are first computed as the solution to a constrained optimization problem. Then the newly computed energy parameters are used to update the constraints on the optimization function, so as to better optimize the energy parameters in the next iteration. Using our method on biologically sound data, we obtain revised parameters for the Turner99 energy model. We show that by using our new parameters, we obtain significant improvements in prediction accuracy over current state of-the-art methods. AVAILABILITY: Our CG implementation is available at http://www.rnasoft.ca/CG/ CONTACT: andrones@cs.ubc.ca. KW - estimation KW - parameter AU - Andronescu, Mirela AU - Condon, Anne AU - Hoos, Holger H AU - Mathews, David H AU - Murphy, Kevin P PY - 2007/07/01/ UR - http://dx.doi.org/10.1093/bioinformatics/btm223 ER -