Suppr超能文献

傅里叶变换红外光谱法(FTIR)中部分氘代后蛋白质二级结构的评估得到改善。

Evaluation of protein secondary structure from FTIR spectra improved after partial deuteration.

机构信息

Center for Structural Biology and Bioinformatics, Laboratory for the Structure and Function of Biological Membranes, Campus Plaine CP206/02, Université Libre de Bruxelles, B1050, Brussels, Belgium.

出版信息

Eur Biophys J. 2021 May;50(3-4):613-628. doi: 10.1007/s00249-021-01502-y. Epub 2021 Feb 3.

Abstract

FTIR spectroscopy has become a major tool to determine protein secondary structure. One of the identified obstacle for reaching better predictions is the strong overlap of bands assigned to different secondary structures. Yet, while for instance disordered structures and α-helical structures absorb almost at the same wavenumber, the absorbance bands are differentially shifted upon deuteration, in part because exchange is much faster for disordered structures. We recorded the FTIR spectra of 85 proteins at different stages of hydrogen/deuterium exchange process using protein microarrays and infrared imaging for high throughput measurements. Several methods were used to relate spectral shape to secondary structure content. While in absolute terms, β-sheet is always better predicted than α-helix content, results consistently indicate an improvement of secondary structure predictions essentially for the α-helix and the category called "Others" (grouping random, turns, bends, etc.) after 15 min of exchange. On the contrary, the β-sheet fraction is better predicted in non-deuterated conditions. Using partial least square regression, the error of prediction for the α-helix content is reduced after 15-min deuteration. Further deuteration degrades the prediction. Error on the prediction for the "Others" structures also decreases after 15-min deuteration. Cross-validation or a single 25-protein test set result in the same overall conclusions.

摘要

傅里叶变换红外(FTIR)光谱已成为确定蛋白质二级结构的主要工具之一。为了提高预测的准确性,目前仍存在一个主要障碍,即不同二级结构所对应的谱带之间存在严重重叠。例如,无规卷曲结构和α-螺旋结构的吸收峰几乎在相同的波数处,但在氘代后,它们的吸收带会发生不同程度的频移,部分原因是无规卷曲结构的交换速度更快。我们使用蛋白质微阵列和红外成像技术,在不同的氢/氘交换阶段对 85 种蛋白质进行了 FTIR 光谱记录,以实现高通量测量。我们使用了几种方法将光谱形状与二级结构含量相关联。虽然从绝对值来看,β-折叠结构的预测总是优于α-螺旋结构,但结果一致表明,在交换 15 分钟后,对α-螺旋和被称为“其他”(包含无规卷曲、转角、弯曲等)的结构的二级结构预测会得到显著改善。相反,在未氘代的条件下,β-折叠结构的预测效果更好。使用偏最小二乘回归(PLSR),在 15 分钟氘代后,对α-螺旋结构含量的预测误差会降低。进一步的氘代会降低预测的准确性。在 15 分钟氘代后,对“其他”结构的预测误差也会降低。交叉验证或单个 25 种蛋白质测试集的结果得出了相同的总体结论。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/411d/8189984/837948cbaa8d/249_2021_1502_Fig1_HTML.jpg

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验