Recent technological advances in mass spectrometry and NMR spectroscopy have enabled new approaches for the rapid and insightful profiling of natural product mixtures. MALDI-MS with the provision of biosynthetic heavy-isotope-labeled precursors can be a powerful method by which to interrogate a natural product metabolome and to gain insight into its unique constituents; this is illustrated herein by the detection, isolation, and characterization of cryptomaldamide. MS/MS-based Molecular Networks, facilitated by the Global Natural Products Social (GNPS) platform, is rapidly changing the way in which we dereplicate known natural products in mixtures, find new analogues in desired structure classes, and identify fundamentally new chemical entities. This method can be linked to genomic information to assist in genome-driven natural products discovery and is illustrated here with the characterization of the columbamides. Similarly, algorithmic interpretation of NMR data is facilitating the automatic identification or classification of new natural products. We developed such a tool named the Small Molecule Accurate Recognition Technology (SMART), which employs a convolutional neural network to classify HSQC spectra of organic molecules using pattern recognition principles. The discovery and rapid classification of several new peptides from a marine cyanobacterium as members of the viequeamide class provides an example of its utility in natural products research. These three illustrations represent different methods by which to look at the external features of a chemical substance and derive valuable insights into its identity or, as described herein, the “face of a molecule”.

中文翻译：

---

分子的脸

质谱和NMR光谱技术的最新技术进步为快速，有洞察力地分析天然产物混合物提供了新的方法。提供生物合成的重同位素标记的前体的MALDI-MS可能是一种强大的方法，可以通过该方法来查询天然产物的代谢组并了解其独特的组成；本文通过隐麦芽糖酰胺的检测，分离和表征来说明这一点。在全球天然产物社会（GNPS）平台的推动下，基于MS / MS的分子网络正在迅速改变我们在混合物中去复制已知天然产物，寻找所需结构类别的新类似物并从根本上鉴定新化学实体的方式。该方法可以与基因组信息联系起来，以协助基因组驱动的天然产物的发现，并在本文中通过哥伦酰胺的表征进行了说明。同样，NMR数据的算法解释也有助于自动识别或分类新的天然产物。我们开发了一种名为小分子精确识别技术（SMART）的工具，该工具使用卷积神经网络使用模式识别原理对有机分子的HSQC光谱进行分类。来自海洋蓝细菌的几种新肽的发现和快速分类（作为维克酰胺类的成员）提供了其在天然产物研究中的效用的一个例子。