Comprehensive Strategy for Proton Chemical Shift Prediction: Linear Prediction with Nonlinear Corrections

详细信息    查看全文

• 作者：Reino Laatikainen ; Tommi Hassinen ; Juuso Lehtivarjo ; Mika Tiainen ; Juha Jungman ; Tuulia Tynkkynen ; Samuli-Petrus Korhonen ; Matthias Niemitz ; Pekka Poutiainen ; Olli J盲盲skel盲inen ; Topi V盲is盲nen ; Janne Weisell ; Pasi Soininen ; Pekka Laatikainen ; Henri Martonen ; Kari Tuppurainen
• 刊名：Journal of Chemical Information and Modeling
• 出版年：2014
• 出版时间：February 24, 2014
• 年：2014
• 卷：54
• 期：2
• 页码：419-430
• 全文大小：375K
• 年卷期：v.54,no.2(February 24, 2014)
• ISSN：1549-960X

文摘

A fast 3D/4D structure-sensitive procedure was developed and assessed for the chemical shift prediction of protons bonded to sp<sup>3sup>carbons, which poses the maybe greatest challenge in the NMR spectral parameter prediction. The LPNC (Linear Prediction with Nonlinear Corrections) approach combines three well-established multivariate methods viz. the principal component regression (PCR), the random forest (RF) algorithm, and the k nearest neighbors (kNN) method. The role of RF is to find nonlinear corrections for the PCR predicted shifts, while kNN is used to take full advantage of similar chemical environments. Two basic molecular models were also compared and discussed: in the MC model the descriptors are computed from an ensemble of the conformers found by conformational search based on Metropolis Monte Carlo (MMC) simulation; in the 4D model the conformational space was further expanded to the fourth dimension (time) by adding molecular dynamics to the MC conformers. An illustrative case study about the application and interpretation of the 4D prediction for a conformationally flexible structure, scopolamine, is described in detail.