High fidelity sampling schedules for NMR spectra of high dynamic range.
High fidelity sampling schedules for NMR spectra of high dynamic range.
复制标题
DOI：
10.1016/j.jmr.2022.107228
复制
发表时间：
2022-06
影响因子：
2.2
通讯作者：
Wagner, Gerhard
中科院分区：
文献类型：
--
作者：
Hyberts, Sven G.;Wagner, Gerhard
关键词：
重建非均匀采样（NUS）NMR光谱的能力大多被接受。仍然担心的是，对从不均匀采样的伪影徘徊。经验丰富，某些抽样时间表比其他采样时间表更好。对于低动态范围频谱和保守的稀疏性，找到有用的时间表是相对微不足道的，但是当动态范围较大并且/或使用极端稀疏性时，并非如此。高动态范围通常在代谢物的Noesy和Spastra中发现，在高保真度时需要对峰高的定量进行定量。当高吞吐量是一个目标时，需要极端稀疏性。在所有情况下，选择较差的采样时间表都可以创建不必要的工件。实际上，重要的是选择一个提供信噪比Apex比率（SAAR）的采样时间表，其标准级或比信噪比（SNR）更好。值得注意的是，通过信噪比，我们将重建保真度视为顶点强度相似性，即作为最高伪像的真实信号。我们表明，重建的质量取决于特定的采样时间表。我们评估了匹配的洛伦兹 - 高斯变换以及常见的瞬息万变和傅立叶变换后，评估了频域中的重建质量。随着Lorentz-to-gauss变换改善了分辨率并减少了山脊，我们在定义信噪比Apex比率（SAAR）度量时包括了这一点。该度量标准衡量模拟重建的峰高度与没有噪声的光谱中最高的重建伪像的比率。一旦找到了最佳SAAR的NUS时间表，对于使用相同参数集录制的所有光谱都将令人满意。补品中提供了良好种子值的表。
The ability to reconstruct non-uniformly sampled (NUS) NMR spectra has mostly been accepted. Still a concern is lingering regarding artifacts from sampling non-uniformly. As experienced, some sampling schedules yield better results than others. Finding a useful schedule is relatively trivial for a low dynamic range spectrum and a conservative sparsity, but not so when the dynamic range is large and/or when extreme sparsity is used. High dynamic range is typically found in NOESY and spectra of metabolites, where quantification of peak heights is desired at high fidelity. Extreme sparsity is desired when high throughput is a goal. In all cases, selecting a poor sampling schedule can create unnecessary artifacts. Effectively, it is important to select a sampling schedule that provides a signal-to-artifact apex ratio (SAAR) to be in par or better than the signal-to-noise ratio (SNR). Notably, by signal-to-artifact apex ratio we consider reconstruction fidelity as the apex intensity likeness, i.e., as the true signal to the tallest artifact. We show that the quality of reconstruction depends on the particular sampling schedule. We evaluate the reconstruction quality in the frequency domain following a matched Lorentz-to-Gauss transform plus common apodization and Fourier Transform. As the Lorentz-to-Gauss transform improves resolution and reduces ridges we include this when defining the Signal-to-Artifact Apex Ratio (SAAR) metric. This metric measures the ratio of simulated reconstructed peak height to the tallest artifact of reconstruction in a spectrum without noise. Once a NUS schedule is found with an optimal SAAR it will be satisfactory for all spectra recorded with the same parameter set. Tables with good seed values are provided in the supplement.
登录
查看更多内容
影响因子：
15
作者：
Hyberts, Sven G.;Takeuchi, Koh;Wagner, Gerhard
通讯作者：
Wagner, Gerhard
影响因子：
2.7
作者：
Nichols, Parker J.;Born, Alexandra;Henen, Morkos A.;Strotz, Dean;Jones, David N.;Delaglio, Frank;Vogeli, Beat
通讯作者：
Vogeli, Beat
影响因子：
2.2
作者：
Hyberts, Sven G.;Arthanari, Haribabu;Wagner, Gerhard
通讯作者：
Wagner, Gerhard
影响因子：
3
作者：
Donoho, DL
通讯作者：
Donoho, DL
影响因子：
2.7
作者：
Hyberts SG;Robson SA;Wagner G
通讯作者：
Wagner G