Modeling the relative response factor of small molecules in positive electrospray ionization

Area of Science:

• Analytical Chemistry
• Computational Chemistry
• Biochemistry

Background:

• Liquid chromatography-electrospray ionization-high-resolution mass spectrometry (LC-ESI-HRMS) is vital for non-targeted analysis (NTA) of complex samples.
• A key challenge in NTA is the scarcity of analytical standards for many detected compounds.
• Computational methods can estimate compound concentrations by modeling relative response factors (RRFs).

Purpose of the Study:

• To develop and evaluate a computational workflow for predicting RRF using molecular dynamics (MD).
• To address the limitation of unavailable analytical standards in NTA of environmental and biological samples.

Main Methods:

• Utilized molecular dynamics (MD) simulations with the CGenFF force field and GROMACS.
• Calculated Lennard-Jones and Coulomb interactions within electrospray ionization (ESI) droplets for 48 compounds.
• Developed a multilinear regression model using Monte Carlo simulations to predict RRF.

Main Results:

• The best predictive model achieved a coefficient of determination (R ²) of 0.82.
• The model demonstrated a mean absolute error (MAE) of 0.13 log units.
• Performance was comparable to existing machine learning models.

Conclusions:

• The developed MD-based computational workflow shows promise for predicting RRFs.
• This approach can aid in quantifying compounds in NTA when analytical standards are unavailable.
• Further validation across diverse chemical structures is warranted for broader applicability.