|
Mark Bayliss, Vitaly Lashin, Antony Williams, Kevin Owens
As MS instrumentation continues to develop enabling the acquisition of yet more data, often from on-the-fly instrument driven MS to MS/MS or MSn experiments, the challenge facing us within the industry then becomes one of extraction, identification of the chromatographic peaks that represent either metabolic activity of the dosed drug compound or impurity related spectra. The volume of data that then requires elucidation to determine structure further challenges the ability to increase sample throughput which continues to be a driving requirement within the industry.
In metabolism studies the vast amounts of MS, MS/MS and MS(n) data are often acquired as a combination of endogenous and metabolite or impurity related data. Extraction and realization of the metabolite or impurity related MS data is further complicated by the extremes in dynamic range of these components and by high level background ions arising from solvents and buffers.
We will report on advances to the Component Detection Algorithm (CODA) algorithm, developed by Windig et al which include advanced baselining and peak selection criteria. The detection of peaks eluting in regions of chromatographic baseline disturbance due to solvent and buffer effects within a gradient elution are also significantly improved as a result. The presence of noise spikes resulting from electronic and spray related events are effectively excluded from the extracted peak information though the use of peak detection methods which in combination with the advanced baselining algorithm has lead to an increased confidence of peak extraction.
During this presentation we will also discuss the impact of using correlation analysis to distinguish metabolite related spectra from endogenous spectra such that the volume of spectra that require structural elucidation may be reduced to those which are most likely metabolite or impurity related. Whilst a mass differential relative to the dosed or stress tested drug can account for the determination of metabolic or degradation activity, enabling the identification of actual structure requires the correlation of known spectral features to known fragmentation events. The usage of spectral correlation is also discussed a means of identification of potential structural modifications. | 
