17th International Mass Spectrometry Conference :: Prague, 2006
> Go to contents (site navigation)
|Session:||High Throughput Analysis: From Discovery to Clinical Applications|
|Presentation date:||Mon, Aug 28, 2006|
|Presentation time:||09:50 – 11:20|
Frank E. Kuhlmann1, Lee H. Altmayer11 Agilent Technologies, Santa Clara, United States
Correspondence address: Frank E. Kuhlmann, Agilent Technologies, 5301 Stevens Creek Blvd, Santa Clara, California, 95052 United States.
Keywords: Automation; MS/MS, Liquid Chromatography; Quadrupole, Triple; Quantitative Analysis.
Novel aspect: Non-derivative based parameter-free integrator for triple quad MRM data with integrated peak quality validation.
Chromatographic peak integration is a crucial step in triple quad MRM quantitation and has a significant impact on the time required for data review. Traditional integrators based on derivative approaches are often difficult to set up for proper integration across the dynamic range, require smoothing of the original chromatograms, perform poorly when approaching the limit of detection, and require time consuming manual re-integration.
We describe a novel automatic, parameter-free integration algorithm specifically constructed and optimized for MRM triple quad data, which does not use noise-amplifying derivatives and peak-distorting filtering. In addition, validation routines determine the quality of each integrated peak and designate it as accepted, questionable or rejected based on different criteria.
The performance of this new integrator is demonstrated by applying it to several sets of typical triple quad MRM data, and comparing it to several derivative-based integrators with regards to unacceptable integration of real peaks (peak start, peak end, and baseline positioning), improper detection of spikes and noise as peaks, the minimum achievable limit of quantitation without manual re-integration, and the total time required for batch review and re-processing. The test data exhibits varying noise, chromatographic peak widths and signal intensities across the typical linear dynamic range of a triple quad MS and include samples of THC, Cocaine, Amphetamine, Metamphetamine and MDMA in biological fluids and several herbicides in soil.
The traditional derivative based integrators often divide the main peak into several peaks, and produce numerous peaks from spikes and disturbances on the baseline toward the lower limit of detection. This requires the user typically to supervise the integration and re-process after adjusting integration parameters or conduct time-consuming manual re-integration.
The new integrator is able to reliably distinguish spikes from peaks in unsmoothed data and ignore the spikes during integration. When encountering merged peaks such as for diastereoisomeric drugs and their metabolites the new integrator was able to properly calculate peak areas, but started to flag peaks as to be inspected or rejected, if the merge valley exceeded certain threshold values. The new integration algorithm also evaluates each integrated peak based on peak height relative to the noise, peak width, peak symmetry and the degree of noise on the peak itself, and suggests inspecting them or rejects them.
In all cases where the new integrator accepted the integration of the compound peaks, they were properly integrated down to concentrations below the lower limit of quantitation. This allowed focusing on inspecting only peaks in question, which resulted in average time savings of about 60 – 90 min. per 1000 integrated compound peaks in the data review of quantitation batches, depending on the user’s skill level.