MSc Thesis: Chemical Space Coverage in LC-HRMS

Literature Thesis Presentation

Investigating the chemical space coverage of liquid chromatographic techniques: how selectivity drives measurability

---

Photo by Nastuh Abootalebi on Unsplash

Literature Thesis Presentation

Investigating the chemical space coverage of liquid chromatographic techniques: how selectivity drives measurability

---

Photo by Nastuh Abootalebi on Unsplash

MSc Literature Thesis: Chemical Space Coverage in LC-HRMS

Investigating the chemical space coverage of liquid chromatographic techniques: how selectivity drives measurability

• Introduction & Problem Statement
• Methodology: RepoRT Analysis
• Chemical Space Coverage Results
• Conclusions & Perspectives

---

Photo by Nastuh Abootalebi on Unsplash

• Introduction: Definition of chemical space and NTA challenges
• Methodology: Data acquisition and instrumental setups
• Results and Analysis: RepoRT database and chemical trends analysis
• Conclusion: Final conclusions and perspectives

---

Photo by Tom Parkes on Unsplash

• Overview of Chemical Space and Non-Targeted Analysis: Introduction to exposome and analytical challenges
• Methodology and Instrumental Setups: Analysis of RepoRT repository and methodology
• Analysis of Chemical Coverage: Chemical coverage, trends, and limitations
• Conclusions and Perspectives: Final summary and future directions

1

Introduction

Chemical space, non-targeted analysis, and the selectivity-measurability bias

---

Photo by Nastuh Abootalebi on Unsplash

• The exposome encompasses all chemical and non-chemical exposures.
• Non-targeted analysis (NTA) with LC-HRMS aims to identify thousands of compounds.
• Selectivity-measurability bias: analytical methods define which chemicals are detectable.
• The study aims to investigate how LC experimental conditions capture different regions of the chemical space.

• Exposome represents the totality of lifelong exposures.
• Non-targeted analysis (NTA) + LC-HRMS is essential for identifying thousands of unknown compounds.
• The "Selectivity-Measurability" bias: Analytical choices dictate which compounds are detected.
• Challenge: Most existing workflows are narrow, potentially missing important chemicals (CECs).

• The exposome includes all chemical and non-chemical exposures throughout a lifetime.
• Non-targeted analysis (NTA) with LC-HRMS is essential for identifying thousands of compounds.
• Selectivity-measurability bias: analytical methods only detect compounds they interact with effectively.
• The thesis investigates if current LC conditions truly cover the relevant chemical space.

• Data source: RepoRT repository (236 curated LC methods).
• Instrumentation: Primarily RPLC (89%) vs. HILIC (11%).
• Reported compounds: 75,797 compounds from 17,083 unique structures.
• Methodological limitations: Lack of extreme polarity/hydrophobicity coverage.

2

Methodology

Leveraging the RepoRT repository for data-driven assessment

---

Photo by Tom Parkes on Unsplash

Reversed-Phase LC (RPLC) Dominates 89% of setups. Excellent for mid-polar to hydrophobic compounds. Often shows reduced detectability for highly polar analytes.

HILIC Separation Represents 11% of methods. Targeted at highly polar and hydrophilic compounds. Currently underrepresented in standard workflows.

• 236: Methods
• 75,797: Compounds
• 89%: RPLC Dominance

• RepoRT repository: 236 curated LC methods using USP-classified columns.
• Data processing used Julia 1.11.4; PubChemCrawler for chemical descriptors.
• Key variables: column.usp.code, length, ID, particle size, temp, flowrate.
• Analysis techniques: Principal Component Analysis (PCA) and k-means clustering.

• Most setups (89%) use RPLC columns, favoring hydrophobic/moderately polar compounds.
• HILIC methods (11%) are underrepresented, limiting coverage of highly polar analytes.
• General trend: Methods are not optimized for broad-scope NTA, but rather targeted metabolomics.

• 75,796: Analytes
• 0-200: TPSA (Ų)
• -4 to 9: log-P

3

Chemical Coverage Analysis

What does the data reveal about LC performance?

---

Photo by Tom Parkes on Unsplash

• Current methods provide substantial coverage within limited physiochemical regions.
• Significant gap exists for extreme polarity and extreme hydrophobicity.
• Combining diverse chromatographic modes (e.g., GC, SFC, IC) is necessary.
• Comprehensive exposome coverage is far from realistic with current approaches.

• Reported compounds span a wide theoretical domain (TPSA 0-852 Ų, log-P -11 to 26.6).
• Effectively measured space is much narrower: 94% of compounds fall within TPSA 0-200 Ų and log-P -4 to 9.
• Clear lack of orthogonal selectivity restricts exploration of extreme chemical regions.
• Current methods offer substantial coverage but remain highly constrained.

Thesis Conclusion: Need for wider scope strategies in NTA

Thank you for your attention. Questions?

---

Photo by Nastuh Abootalebi on Unsplash

• 236: LC Methods
• 75,797: Compounds
• 89%: RPLC Usage
• 11%: HILIC Usage

Complete chemical space coverage is not currently realistic with existing methods.

Advancing toward comprehensive exposome monitoring requires more than current LC workflows.

• Dominance of RPLC (89%) indicates bias toward non-polar to moderately polar compounds.
• The most frequent covered chemical domain is narrow (TPSA 0-200 Å2; log-P -4 to 9).
• Lack of orthogonal selectivity limits exploration of extreme physicochemical regions.
• Large overlap between RPLC and HILIC methods in targeted applications.

• Analyze exclusively NTA-generated datasets.
• Increase chromatographic diversity (e.g., incorporate GC, SFC, IC).
• Include more detailed metadata (e.g., eluent pH, gradient profiles) for better modeling.
• Goal: Move beyond LC-based constraints for true exposome characterization.

Conclusions & Perspectives

The journey toward comprehensive chemical space monitoring continues.

---

Photo by Nastuh Abootalebi on Unsplash

• Current LC methods are substantial but constrained; complete coverage is unrealistic under current conditions.
• Systematic under-detection of highly polar and extremely hydrophobic compound classes.
• Future work: focus on exclusive NTA datasets and broader method variety (GC, SFC, IC).
• Essential to combine orthogonal techniques for a true exposome representation.