INTRODUCTION Marine planktonic communities are complex microbial consortia often dominated by microscopic algae. The taxonomic identification of individual phytoplankton cells usually relies on their morphology and demands expert knowledge. Recently, a live single-cell mass spectrometry (LSC-MS) pipeline was developed to generate metabolic profiles of microalgae. OBJECTIVE Taxonomic identification of diverse microalgal single cells from collection strains and plankton samples based on the metabolic fingerprints analyzed with matrix-free laser desorption/ionization high-resolution mass spectrometry. METHODS Matrix-free atmospheric pressure laser-desorption ionization mass spectrometry was performed to acquire single-cell mass spectra from collection strains and prior identified environmental isolates. The computational identification of microalgal species was performed by spectral pattern matching (SPM). Three similarity scores and a bootstrap-derived confidence score were evaluated in terms of their classification performance. The effects of high and low-mass resolutions on the classification success were evaluated. RESULTS Several hundred single-cell mass spectra from nine genera and nine species of marine microalgae were obtained. SPM enabled the identification of single cells at the genus and species level with high accuracies. The receiver operating characteristic (ROC) curves indicated a good performance of the similarity measures but were outperformed by the bootstrap-derived confidence scores. CONCLUSION This is the first study to solve taxonomic identification of microalgae based on the metabolic fingerprints of the individual cell using an SPM approach.

中文翻译：

---

使用无基质激光/解吸电离质谱法从浮游生物样品中鉴定活的单个微藻细胞的物种水平。

引言海洋浮游生物群落是复杂的微生物群落，通常以微观藻类为主导。单个浮游植物细胞的分类学鉴定通常依赖于它们的形态并需要专业知识。最近，开发了活单细胞质谱（LSC-MS）管道以生成微藻的代谢谱。目的基于代谢指纹图谱的无基质激光解吸/电离高分辨率质谱分析，从收集菌株和浮游生物样品中鉴定各种微藻单细胞的分类学。方法采用无基质大气压激光解吸电离质谱法，从收集菌株和事先鉴定的环境分离物中获得单细胞质谱。微藻物种的计算鉴定是通过光谱模式匹配（SPM）进行的。根据它们的分类性能，评估了三个相似性评分和一个自举的置信度评分。评估了高和低质量分辨率对分类成功的影响。结果获得了来自九属和九种海洋微藻的数百个单细胞质谱。SPM能够以高准确度鉴定属和种水平的单个细胞。接收器工作特性（ROC）曲线表明相似度量的良好性能，但优于自举方法得出的置信度得分。