Combining Ion Mobility Mass Spectrometry and Advanced Predictive Tools for the Structure Determination and Isomer Differentiation.

研究背景质谱（MS）已成为分析不同类型样品的重要方法，从药物代谢物到蛋白质复合物及其结构表征，在从医学到材料科学的广泛领域。在制药工业中，该技术是药品生产过程中质量控制和副产品鉴定的关键，通常作为液相色谱（LC）的多维方法。离子迁移率（IM，一种气相分离技术）已经改变了分离和表征复杂混合物中分子的能力，特别是异构体形式（具有相同质量）。它根据离子的旋转平均三维结构（碰撞截面）来分离离子，因此可用于异构体化合物的分离。目前还没有严格的框架，如何这样的CCS值应记录在不同的仪器，记录和存储在可搜索的数据库，这严重限制了使用离子迁移率数据forcompound identification和data mining.aims和objectivesThe目标是：在这里，我们开发了一个数据标准的离子迁移率的小分子，这将使自动识别和AI方法，以利用额外的分析层面，离子迁移率可以提供。我们将开发机器学习来解释多维数据并预测分子结构。我们将使用良好表征的化合物的训练集来系统地测量离子迁移率特性，例如漂移时间，作为分析工作流程中分子的可能附加标识符，并研究了液相色谱中漂移时间和保留时间之间可能存在的相关性。潜在的应用和益处先进的预测工具的使用可能使通过学习不同化合物的模式，更容易解释大的质谱数据集，这可以用于潜在的“填补空白”，以识别特定的片段峰，并可以用于代谢组学和蛋白质组学中的蛋白质修饰。高级预测工具也可以预测其他模式，包括未知物质的碎片模式和保留时间。研究领域：离子迁移质谱，用于结构测定和异构体区分的高级预测工具。资格要求：博士学位

Context of researchMass spectrometry (MS) has become an essential approach for analysis of different types of samples, from drugmetabolites to protein complexes and their structural characterization, in a wide range of fields from medicine to materialsscience. In the pharmaceutical industry the technique is key for quality control and identification of byproducts in drugmanufacturing, usually as a multidimensional approach with liquid chromatography (LC).Ion mobility (IM, a gas-phase separation technique) has transformed capabilities to separate and characterize molecules incomplex mixtures, particularly isomeric forms (with the same mass). It separates ions in relation to their roationallyaveraged three-dimensional structure (collision cross section, ccs), and therefore can be useful for the separation ofisomeric compounds. There is currently no stringent framework how such ccs values should be recorded across differentinstruments, documented and deposited in searchable databases, which severely limits the use of ion mobility data forcompound identification and data mining.Aims and objectivesThe objectives are:Here we develop a data standard for ion mobility of small molecules which will enable automated identification and AIapproaches to take advantage of the additional analytical dimension which ion mobility can provide. We will developmachine learning to interpret multi-dimensional data and predict molecular structures.We will use training sets of wellcharacterizedcompounds to systematically measure ion mobility properties such as the drift time, as a possible additionalidentifier for molecules in analytical workflows, and investigate possible correlations between drift times and retention timesin liquid chromatography.Potential applications and benefitsThe use of advance prediction tools may make interpreting large mass spectrometry data sets easier via learning patternsin different compounds, this can be used to potentially "fill-in-gaps" to identify specific fragment peaks and may be used inmetabolomics and proteomics to protein modifications. Advanced predictive tool may also predict other patterns includingfragmentation patterns and retention times of unknown substances.Research Areas: Ion Mobility Mass Spectrometry, Advanced Predictive Tools for the Structure Determination and IsomerDifferentiation.Qualification to be attained: PhD degree