Estimation of low-level components lost through chromatographic separations with finite detection limits

Area of Science:

• Analytical Chemistry
• Biomarker Discovery
• Chromatography

Background:

• Liquid chromatography (LC) is crucial for identifying disease biomarkers.
• Significant challenges exist in detecting low-concentration components due to detection limits and peak overlap.
• Previous studies explored signal intensity distributions but lacked comprehensive analysis of loss mechanisms.

Purpose of the Study:

• To investigate the extent of component loss in LC, considering both limit of detection (LOD) and peak overlap.
• To compare different probability density functions (PDFs) for signal intensity distribution, including log-normal, Weibull, and exponential.
• To simulate chromatograms and assess the impact of column efficiency and signal-to-noise ratio (SNR) on component loss.

Main Methods:

• Extended previous work on missing components using log-normal, Weibull, and exponential PDFs.
• Simulated synthetic chromatograms to compare peak loss from overlap versus LOD.
• Analyzed component loss as a function of relative column efficiency (saturation) and SNR.

Main Results:

• Curve fitting for low-intensity signal loss assessments proved ambiguous.
• At high column efficiencies, low-amplitude peaks are primarily lost below the LOD.
• At lower efficiencies, peak overlap emerges as the dominant mechanism for component loss.

Conclusions:

• Single-channel detectors in LC struggle to detect low-level components, irrespective of SNR, due to dominant peak overlap at lower efficiencies.
• Multichannel detectors, such as mass spectrometers, are necessary to improve the detection of low-level components.
• Understanding these loss mechanisms is critical for accurate biomarker discovery and disease assessment.