Developing a Scoring Function for NMR Structure-based Assignments using Machine Learning
Determining the assignment of signals received from the ex- periments (peaks) to speci_c nuclei of the target molecule in Nuclear Magnetic Resonance (NMR1) spectroscopy is an important challenge. Nuclear Vector Replacement (NVR) ([2, 3]) is a framework for structure- based assignments which combines multiple types of NMR data such as chemical shifts, residual dipolar couplings, and NOEs. NVR-BIP  is a tool which utilizes a scoring function with a binary integer programming (BIP) model to perform the assignments. In this paper, support vector machines (SVM) and boosting are employed to combine the terms in NVR-BIP's scoring function by viewing the assignment as a classi_ca- tion problem. The assignment accuracies obtained using this approach show that boosting improves the assignment accuracy of NVR-BIP on our data set when RDCs are not available and outperforms SVMs. With RDCs, boosting and SVMs o_er mixed results.
Unable to display preview. Download preview PDF.
- 1.Apaydin, M. S., Çatay, B., Patrick N. and Donald, B. R.: NVR-BIP: Nuclear Vector Replacement using Binary Integer Programming for NMR Structure-Based Assign- ments. The Computer Journal, Advance Access published on January 6, 2010; doi: doi:10.1093/comjnl/bxp120.Google Scholar
- 2.Langmead, C., Yan, A., Lilien, R., Wang, L., and Donald, B.: A Polynomial-Time Nuclear Vector Replacement Algorithm for Automated NMR Resonance Assignments (2003) In Proc. The Seventh Annual International Conference on Research in Computational Molecular Biology (RECOMB) Berlin, Germany, April 1013: ACM Press. appears in: J. Comp. Bio. (2004), 11 (2–3), pp. 277–98 pp. 176–187.Google Scholar