Related Experiment Video
Updated: Jul 3, 2026

Using a Cyclic Ion Mobility Spectrometer for Tandem Ion Mobility Experiments
Published on: January 20, 2022
Interpretable machine learning-based automated HPLC/MS2 platform using ion-molecule reactions for the identification
Armen G Beck1, Ruth O Anyaeche1, Prageeth Wijewardhane1
1Department of Chemistry, Purdue University 560 Oval Drive West Lafayette IN USA hilkka@purdue.edu gchopra@purdue.edu.
We developed an interpretable machine learning method for automated compound identification in complex mixtures. This approach uses chemical graphs to differentiate similar ions, advancing automated chemical analysis.
Area of Science:
- Analytical Chemistry
- Computational Chemistry
Background:
- Compound identification in complex mixtures is challenging using traditional methods like High-Performance Liquid Chromatography (HPLC) coupled to tandem mass spectrometry (MS2) with collision-activated dissociation (CAD).
- CAD often yields similar fragmentation patterns for isomeric or related compounds, hindering differentiation.
- Gas-phase ion-molecule reactions offer an alternative for differentiating ions by identifying functional groups, but manual interpretation and optimization are complex.
Purpose of the Study:
- To develop an automated, interpretable machine learning approach for identifying unknown compounds and differentiating similar ions in complex mixtures.
- To enable automated selection of reagents and optimization of reaction conditions for ion-molecule reactions in mass spectrometry.
- To lay the groundwork for fully automated HPLC/MS2 platforms for chemical analysis.
Main Methods:
- A chemical graph-based interpretable machine learning approach was developed.
- The method automates the identification of functionalities in protonated analytes.
- It enables automated selection of neutral reagents and optimization of reagent introduction parameters (pulsing-in and pumping-out times) for mass spectrometry.
Main Results:
- The developed approach successfully automates the identification of functionalities in unknown protonated analytes.
- It facilitates the differentiation of isomeric or otherwise similar compounds.
- This method allows for the automated selection of reagents and optimization of reaction conditions, overcoming previous manual limitations.
Conclusions:
- The chemical graph-based machine learning approach provides a powerful tool for automated compound identification and differentiation.
- This work advances ion-molecule reaction studies by enabling automated reagent selection and condition optimization.
- The findings establish a foundation for fully automated HPLC/MS2 platforms with broad applications in chemical sciences.
Related Concept Videos
High-Performance Liquid Chromatography: Types of Detectors
Mass Spectrometry: Complex Analysis
GC–MS is a powerful hyphenated method commonly used in forensics and environmental...
Peptide Identification Using Tandem Mass Spectrometry
This technique helps gather information regarding the protein from which the peptide was obtained and to study the peptides’ amino acid sequence. Identifying peptides from a complex mixture is an important component of the growing field of...
Ion-Exchange Chromatography
High-Resolution Mass Spectrometry (HRMS)
High-Performance Liquid Chromatography: Instrumentation

