Department of Forensic Medicine, Faculty of Health and Medical Sciences, Section of Forensic Chemistry, University of Copenhagen, Copenhagen, Denmark.
PLoS One. 2020 Nov 12;15(11):e0242224. doi: 10.1371/journal.pone.0242224. eCollection 2020.
The ever-changing market of new psychoactive substances (NPS) poses challenges for laboratories worldwide. Analytical toxicologists are constantly working to keep high-resolution mass spectrometry (HR-MS) screening libraries updated for NPS. This study sought to use the online crowd-sourced HighResNPS database for spectrum comparison screening, thereby broadening its utility to all HR-MS instruments.
HighResNPS allows formation of a set of consensus fragment ions for a NPS and prioritises among multiple entries of collision-induced fragment ions. A subset of 42 NPS samples was analysed in data-independent acquisition (DIA) and data-dependent acquisition (DDA) modes on two different instruments. HighResNPS-computed spectra were generated with either Absolute (all fragment ions set to 100%) or Fractional (50% intensity reduction of former fragment ion) intensity. The acquired NPS data were analysed using the consensus library with computed ion intensities and evaluated with vendor-neutral screening software.
Overall, of the 42 samples, 100% were identified, with 88% identified as the top candidate. Three samples had the correct candidate proposed as the second highest ranking NPS. In all three of those samples, the top proposed candidate was a positional isomer or closely related compound. Absolute intensity assignment provided identical scoring between the top two proposed compounds in two samples with DIA. DDA had a slightly higher identification rate in the spectra comparison screening with fractional intensity assignment, but no major differences were observed.
The fractional intensity assignment was slightly more advantageous than the absolute assignment. It was selective between proposed candidates, showed a high identification rate and had an overall higher fragmentation scoring. The candidates proposed by the HighResNPS library spectra comparison simplify the determination of NPS for researchers and toxicologists. The database provides free monthly updates of consensus spectra, thereby enabling laboratories to stay at the forefront of NPS screening by LC-HR-MS with spectra screening software.
新精神活性物质（NPS）不断变化的市场给世界各地的实验室带来了挑战。分析毒理学家一直在努力更新高分辨质谱（HR-MS）筛查库，以涵盖 NPS。本研究旨在使用在线众包 HighResNPS 数据库进行谱比较筛选，从而扩大其在所有 HR-MS 仪器中的应用。
HighResNPS 允许为 NPS 形成一组共识碎片离子，并在多个碰撞诱导碎片离子条目之间进行优先级排序。在两台不同的仪器上，以数据非依赖性采集（DIA）和数据依赖性采集（DDA）模式分析了一组 42 个 NPS 样本。HighResNPS 计算的光谱使用绝对（所有碎片离子设置为 100%）或分数（前一个碎片离子强度降低 50%）强度生成。使用共识库和计算的离子强度分析采集的 NPS 数据，并使用供应商中立的筛选软件进行评估。
总体而言，42 个样本中有 100%被鉴定，其中 88%被鉴定为首选候选。有 3 个样本中正确的候选物被提议为排名第二的 NPS。在这 3 个样本中，首选候选物均为位置异构体或密切相关的化合物。在使用 DIA 的两个样本中，绝对强度赋值为前两个提议化合物提供了相同的评分。在使用分数强度赋值的谱比较筛选中，DDA 的鉴定率略高，但未观察到重大差异。
分数强度赋值比绝对赋值略有优势。它在提议候选物之间具有选择性，显示出高鉴定率和整体更高的碎片评分。HighResNPS 库谱比较提出的候选物简化了研究人员和毒理学家对 NPS 的确定。该数据库每月免费提供共识谱的更新，从而使实验室能够通过 LC-HR-MS 和谱筛选软件保持在 NPS 筛选的前沿。
