A principal component analysis approach for developing retention models in liquid chromatography.

Area of Science:

• Analytical Chemistry
• Chromatography
• Chemometrics

Background:

• Liquid chromatography (LC) retention modeling is crucial for predicting solute behavior.
• Linear solvation energy relationship (LSER) models are commonly used but have limitations.
• Principal component analysis (PCA) offers an alternative multivariate approach.

Purpose of the Study:

• To develop and evaluate novel retention models for liquid chromatography using PCA.
• To compare the performance of PCA-based models against traditional LSER models.
• To assess the predictive capability of these models for novel chromatographic conditions.

Main Methods:

• Development of three retention models utilizing principal component analysis (PCA).
• Comparison of PCA model fitting performance with a linear solvation energy relationship (LSER) model.
• Evaluation of model performance using artificial neural networks (ANNs).
• Testing predictive accuracy across diverse analyte datasets (non-polar to polar compounds).

Main Results:

• PCA models demonstrated comparable features to LSER models.
• PCA models exhibited superior fitting performance compared to the standard LSER model.
• Artificial neural networks significantly enhanced the performance of the LSER model.
• Models showed potential for predicting retention times for previously unstudied solutes and conditions.

Conclusions:

• PCA provides a robust framework for developing effective liquid chromatography retention models.
• PCA-based models offer advantages in fitting performance and predictive power.
• The integration of ANNs can further optimize LSER model predictions.
• These models hold promise for advancing solute retention prediction in chromatography.